From 0dd3ab0d81f67506547136532af39b5c9a16771e Mon Sep 17 00:00:00 2001
From: Matthias Mohr <m.mohr@uni-muenster.de>
Date: Thu, 10 Nov 2022 16:30:09 +0100
Subject: [PATCH 01/52] Deploy a separate draft branch for 2.0

---
 .github/workflows/docs.yml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.github/workflows/docs.yml b/.github/workflows/docs.yml
index 260743f4..8ac28b02 100644
--- a/.github/workflows/docs.yml
+++ b/.github/workflows/docs.yml
@@ -5,6 +5,7 @@ on:
   push:
     branches:
       - draft
+      - draft-2.0
       - master
 jobs:
   deploy:

From 49372904bc2d46c38c7d82ec1d4b5b4accbf914c Mon Sep 17 00:00:00 2001
From: Matthias Mohr <webmaster@mamo-net.de>
Date: Fri, 26 Jul 2024 19:53:10 +0200
Subject: [PATCH 02/52] Update docgen

---
 tests/docs.html | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/tests/docs.html b/tests/docs.html
index 23ef98e4..981ed3ab 100644
--- a/tests/docs.html
+++ b/tests/docs.html
@@ -5,7 +5,7 @@
     <meta http-equiv="x-ua-compatible" content="ie=edge">
     <title>openEO API Processes</title>
     <link rel="stylesheet" href="https://fonts.googleapis.com/css?family=Roboto:300,400,400i,700|Roboto+Mono">
-    <script src="https://cdn.jsdelivr.net/npm/vue@2.6"></script>
+    <script src="https://cdn.jsdelivr.net/npm/vue@2.7"></script>
     <script src="https://cdn.jsdelivr.net/npm/@openeo/processes-docgen@1/dist/DocGen.umd.min.js"></script>
     <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/@openeo/processes-docgen@1/dist/DocGen.css">
     <style>
@@ -115,7 +115,9 @@
           categorize: true,
           apiVersion: '1.1.0',
           title: 'openEO processes (draft)',
-          notice: '**Note:** This is the list of all processes specified by the openEO project. Back-ends implement a varying set of processes. Thus, the processes you can use at a specific back-end may derive from the specification, may include non-standardized processes and may not implement all processes listed here. Please check each back-end individually for the processes they support. The client libraries usually have a function called `listProcesses` or `list_processes` for that.'
+          notice: '**Note:** This is the list of all processes specified by the openEO project. Back-ends implement a varying set of processes. Thus, the processes you can use at a specific back-end may derive from the specification, may include non-standardized processes and may not implement all processes listed here. Please check each back-end individually for the processes they support. The client libraries usually have a function called `listProcesses` or `list_processes` for that.',
+          showExperimentalByDefault: true,
+          showDeprecatedByDefault: false,
         }
       })
     });

From 6141771129d58d2a292db3b91d80c812c49f4e52 Mon Sep 17 00:00:00 2001
From: Michele Claus <31700619+clausmichele@users.noreply.github.com>
Date: Mon, 29 Jul 2024 12:34:31 +0200
Subject: [PATCH 03/52] Remove mention to examples folder (#514)

* Remove mention to examples folder

* Update README.md

---------

Co-authored-by: Matthias Mohr <matthias@mohr.ws>
---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 24c28899..71f78f57 100644
--- a/README.md
+++ b/README.md
@@ -40,7 +40,7 @@ This repository contains a set of files formally describing the openEO Processes
 ## Process
 
 * All new processes must be added to the [`proposals`](proposals/) folder.
-* Processes will only be moved from proposals to the stable process specifications once there are at least two implementations and an example process in the [`examples`](examples/) folder showing it in a use case. This doesn't require a PSC vote individually as it's not a breaking change, just an addition.
+* Processes will only be moved from proposals to the stable process specifications once there are at least two implementations and an example process in the [`openEO community examples`](https://github.com/Open-EO/openeo-community-examples/) showing it in a use case. This doesn't require a PSC vote individually as it's not a breaking change, just an addition.
 * The [`proposals`](proposals/) folder allows breaking changes without a PSC vote and without increasing the major version number (i.e. a breaking change in the proposals doesn't require us to make the next version number 2.0.0).
 * The proposals are released as experimental processes with the other processes.
 * Each release and all breaking changes in the stable process specifications must go through PSC vote.
\ No newline at end of file

From b57cc5bf18f8f2a99321e66ca4693ed5211d84c3 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Mon, 7 Apr 2025 17:08:20 +0200
Subject: [PATCH 04/52] ml proposal

---
 .github/workflows/docs.yml                    |   9 +-
 .../aggregate_spatial_window.json             |   0
 .../apply_polygon.json                        |   0
 .../ard_normalized_radar_backscatter.json     |   0
 .../ard_surface_reflectance.json              |   0
 .../array_create_labeled.json                 |   0
 .../array_find_label.json                     |   0
 .../array_modify.json                         |   0
 .../atmospheric_correction.json               |   0
 .../cloud_detection.json                      |   0
 {proposals => older-proposals}/cummax.json    |   0
 {proposals => older-proposals}/cummin.json    |   0
 .../cumproduct.json                           |   0
 {proposals => older-proposals}/cumsum.json    |   0
 .../date_between.json                         |   0
 .../date_difference.json                      |   0
 .../date_shift.json                           |   0
 .../filter_labels.json                        |   0
 .../filter_vector.json                        |   0
 {proposals => older-proposals}/fit_curve.json |   0
 .../flatten_dimensions.json                   |   0
 {proposals => older-proposals}/inspect.json   |   0
 .../is_infinite.json                          |   0
 .../load_geojson.json                         |   0
 older-proposals/load_ml_model.json            |  46 +++++
 {proposals => older-proposals}/load_stac.json |   0
 .../load_uploaded_files.json                  |   0
 {proposals => older-proposals}/load_url.json  |   0
 .../ml_fit_class_random_forest.json           |   0
 .../ml_fit_regr_random_forest.json            |   0
 older-proposals/ml_predict.json               |  49 ++++++
 {proposals => older-proposals}/nan.json       |   0
 .../predict_curve.json                        |   0
 .../reduce_spatial.json                       |   0
 .../run_udf_externally.json                   |   0
 .../sar_backscatter.json                      |   0
 older-proposals/save_ml_model.json            |  44 +++++
 .../unflatten_dimension.json                  |   0
 .../vector_buffer.json                        |   0
 .../vector_reproject.json                     |   0
 .../vector_to_random_points.json              |   0
 .../vector_to_regular_points.json             |   0
 proposals/ML-README.md                        | 159 ++++++++++++++++++
 proposals/load_ml_model.json                  |   8 +-
 proposals/ml_fit.json                         |  63 +++++++
 proposals/ml_predict.json                     |   6 +-
 proposals/mlm_class_random_forest.json        |  67 ++++++++
 proposals/mlm_regr_random_forest.json         |  67 ++++++++
 proposals/save_ml_model.json                  |   6 +-
 49 files changed, 510 insertions(+), 14 deletions(-)
 rename {proposals => older-proposals}/aggregate_spatial_window.json (100%)
 rename {proposals => older-proposals}/apply_polygon.json (100%)
 rename {proposals => older-proposals}/ard_normalized_radar_backscatter.json (100%)
 rename {proposals => older-proposals}/ard_surface_reflectance.json (100%)
 rename {proposals => older-proposals}/array_create_labeled.json (100%)
 rename {proposals => older-proposals}/array_find_label.json (100%)
 rename {proposals => older-proposals}/array_modify.json (100%)
 rename {proposals => older-proposals}/atmospheric_correction.json (100%)
 rename {proposals => older-proposals}/cloud_detection.json (100%)
 rename {proposals => older-proposals}/cummax.json (100%)
 rename {proposals => older-proposals}/cummin.json (100%)
 rename {proposals => older-proposals}/cumproduct.json (100%)
 rename {proposals => older-proposals}/cumsum.json (100%)
 rename {proposals => older-proposals}/date_between.json (100%)
 rename {proposals => older-proposals}/date_difference.json (100%)
 rename {proposals => older-proposals}/date_shift.json (100%)
 rename {proposals => older-proposals}/filter_labels.json (100%)
 rename {proposals => older-proposals}/filter_vector.json (100%)
 rename {proposals => older-proposals}/fit_curve.json (100%)
 rename {proposals => older-proposals}/flatten_dimensions.json (100%)
 rename {proposals => older-proposals}/inspect.json (100%)
 rename {proposals => older-proposals}/is_infinite.json (100%)
 rename {proposals => older-proposals}/load_geojson.json (100%)
 create mode 100644 older-proposals/load_ml_model.json
 rename {proposals => older-proposals}/load_stac.json (100%)
 rename {proposals => older-proposals}/load_uploaded_files.json (100%)
 rename {proposals => older-proposals}/load_url.json (100%)
 rename {proposals => older-proposals}/ml_fit_class_random_forest.json (100%)
 rename {proposals => older-proposals}/ml_fit_regr_random_forest.json (100%)
 create mode 100644 older-proposals/ml_predict.json
 rename {proposals => older-proposals}/nan.json (100%)
 rename {proposals => older-proposals}/predict_curve.json (100%)
 rename {proposals => older-proposals}/reduce_spatial.json (100%)
 rename {proposals => older-proposals}/run_udf_externally.json (100%)
 rename {proposals => older-proposals}/sar_backscatter.json (100%)
 create mode 100644 older-proposals/save_ml_model.json
 rename {proposals => older-proposals}/unflatten_dimension.json (100%)
 rename {proposals => older-proposals}/vector_buffer.json (100%)
 rename {proposals => older-proposals}/vector_reproject.json (100%)
 rename {proposals => older-proposals}/vector_to_random_points.json (100%)
 rename {proposals => older-proposals}/vector_to_regular_points.json (100%)
 create mode 100644 proposals/ML-README.md
 create mode 100644 proposals/ml_fit.json
 create mode 100644 proposals/mlm_class_random_forest.json
 create mode 100644 proposals/mlm_regr_random_forest.json

diff --git a/.github/workflows/docs.yml b/.github/workflows/docs.yml
index 29baf6db..aff34e08 100644
--- a/.github/workflows/docs.yml
+++ b/.github/workflows/docs.yml
@@ -4,6 +4,7 @@ on:
     types: [published]
   push:
     branches:
+      - dev
       - draft
       - master
 jobs:
@@ -14,7 +15,7 @@ jobs:
         uses: rlespinasse/github-slug-action@v3.x
       - uses: actions/setup-node@v3
         with:
-          node-version: 'lts/*'
+          node-version: "lts/*"
       - uses: actions/checkout@v3
       - run: |
           npm install
@@ -40,7 +41,7 @@ jobs:
         with:
           github_token: ${{ secrets.GITHUB_TOKEN }}
           publish_dir: gh-pages
-          user_name: 'openEO CI'
+          user_name: "openEO CI"
           user_email: openeo.ci@uni-muenster.de
           cname: processes.openeo.org
       - name: deploy to ${{ env.GITHUB_REF_SLUG }}
@@ -50,5 +51,5 @@ jobs:
           github_token: ${{ secrets.GITHUB_TOKEN }}
           publish_dir: gh-pages
           destination_dir: ${{ env.GITHUB_REF_SLUG }}
-          user_name: 'openEO CI'
-          user_email: openeo.ci@uni-muenster.de
\ No newline at end of file
+          user_name: "openEO CI"
+          user_email: openeo.ci@uni-muenster.de
diff --git a/proposals/aggregate_spatial_window.json b/older-proposals/aggregate_spatial_window.json
similarity index 100%
rename from proposals/aggregate_spatial_window.json
rename to older-proposals/aggregate_spatial_window.json
diff --git a/proposals/apply_polygon.json b/older-proposals/apply_polygon.json
similarity index 100%
rename from proposals/apply_polygon.json
rename to older-proposals/apply_polygon.json
diff --git a/proposals/ard_normalized_radar_backscatter.json b/older-proposals/ard_normalized_radar_backscatter.json
similarity index 100%
rename from proposals/ard_normalized_radar_backscatter.json
rename to older-proposals/ard_normalized_radar_backscatter.json
diff --git a/proposals/ard_surface_reflectance.json b/older-proposals/ard_surface_reflectance.json
similarity index 100%
rename from proposals/ard_surface_reflectance.json
rename to older-proposals/ard_surface_reflectance.json
diff --git a/proposals/array_create_labeled.json b/older-proposals/array_create_labeled.json
similarity index 100%
rename from proposals/array_create_labeled.json
rename to older-proposals/array_create_labeled.json
diff --git a/proposals/array_find_label.json b/older-proposals/array_find_label.json
similarity index 100%
rename from proposals/array_find_label.json
rename to older-proposals/array_find_label.json
diff --git a/proposals/array_modify.json b/older-proposals/array_modify.json
similarity index 100%
rename from proposals/array_modify.json
rename to older-proposals/array_modify.json
diff --git a/proposals/atmospheric_correction.json b/older-proposals/atmospheric_correction.json
similarity index 100%
rename from proposals/atmospheric_correction.json
rename to older-proposals/atmospheric_correction.json
diff --git a/proposals/cloud_detection.json b/older-proposals/cloud_detection.json
similarity index 100%
rename from proposals/cloud_detection.json
rename to older-proposals/cloud_detection.json
diff --git a/proposals/cummax.json b/older-proposals/cummax.json
similarity index 100%
rename from proposals/cummax.json
rename to older-proposals/cummax.json
diff --git a/proposals/cummin.json b/older-proposals/cummin.json
similarity index 100%
rename from proposals/cummin.json
rename to older-proposals/cummin.json
diff --git a/proposals/cumproduct.json b/older-proposals/cumproduct.json
similarity index 100%
rename from proposals/cumproduct.json
rename to older-proposals/cumproduct.json
diff --git a/proposals/cumsum.json b/older-proposals/cumsum.json
similarity index 100%
rename from proposals/cumsum.json
rename to older-proposals/cumsum.json
diff --git a/proposals/date_between.json b/older-proposals/date_between.json
similarity index 100%
rename from proposals/date_between.json
rename to older-proposals/date_between.json
diff --git a/proposals/date_difference.json b/older-proposals/date_difference.json
similarity index 100%
rename from proposals/date_difference.json
rename to older-proposals/date_difference.json
diff --git a/proposals/date_shift.json b/older-proposals/date_shift.json
similarity index 100%
rename from proposals/date_shift.json
rename to older-proposals/date_shift.json
diff --git a/proposals/filter_labels.json b/older-proposals/filter_labels.json
similarity index 100%
rename from proposals/filter_labels.json
rename to older-proposals/filter_labels.json
diff --git a/proposals/filter_vector.json b/older-proposals/filter_vector.json
similarity index 100%
rename from proposals/filter_vector.json
rename to older-proposals/filter_vector.json
diff --git a/proposals/fit_curve.json b/older-proposals/fit_curve.json
similarity index 100%
rename from proposals/fit_curve.json
rename to older-proposals/fit_curve.json
diff --git a/proposals/flatten_dimensions.json b/older-proposals/flatten_dimensions.json
similarity index 100%
rename from proposals/flatten_dimensions.json
rename to older-proposals/flatten_dimensions.json
diff --git a/proposals/inspect.json b/older-proposals/inspect.json
similarity index 100%
rename from proposals/inspect.json
rename to older-proposals/inspect.json
diff --git a/proposals/is_infinite.json b/older-proposals/is_infinite.json
similarity index 100%
rename from proposals/is_infinite.json
rename to older-proposals/is_infinite.json
diff --git a/proposals/load_geojson.json b/older-proposals/load_geojson.json
similarity index 100%
rename from proposals/load_geojson.json
rename to older-proposals/load_geojson.json
diff --git a/older-proposals/load_ml_model.json b/older-proposals/load_ml_model.json
new file mode 100644
index 00000000..7fa86d89
--- /dev/null
+++ b/older-proposals/load_ml_model.json
@@ -0,0 +1,46 @@
+{
+    "id": "load_ml_model",
+    "summary": "Load a ML model",
+    "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit_regr_random_forest()`` and ``save_ml_model()``.",
+    "categories": [
+        "machine learning",
+        "import"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "uri",
+            "description": "The STAC Item to load the machine learning model from. The STAC Item must implement the `ml-model` extension.",
+            "schema": [
+                {
+                    "title": "URL",
+                    "type": "string",
+                    "format": "uri",
+                    "subtype": "uri",
+                    "pattern": "^https?://"
+                },
+                {
+                    "title": "User-uploaded File",
+                    "type": "string",
+                    "subtype": "file-path",
+                    "pattern": "^[^\r\n\\:'\"]+$"
+                }
+            ]
+        }
+    ],
+    "returns": {
+        "description": "A machine learning model to be used with machine learning processes such as ``ml_predict()``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://github.com/stac-extensions/ml-model",
+            "title": "STAC ml-model extension",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
diff --git a/proposals/load_stac.json b/older-proposals/load_stac.json
similarity index 100%
rename from proposals/load_stac.json
rename to older-proposals/load_stac.json
diff --git a/proposals/load_uploaded_files.json b/older-proposals/load_uploaded_files.json
similarity index 100%
rename from proposals/load_uploaded_files.json
rename to older-proposals/load_uploaded_files.json
diff --git a/proposals/load_url.json b/older-proposals/load_url.json
similarity index 100%
rename from proposals/load_url.json
rename to older-proposals/load_url.json
diff --git a/proposals/ml_fit_class_random_forest.json b/older-proposals/ml_fit_class_random_forest.json
similarity index 100%
rename from proposals/ml_fit_class_random_forest.json
rename to older-proposals/ml_fit_class_random_forest.json
diff --git a/proposals/ml_fit_regr_random_forest.json b/older-proposals/ml_fit_regr_random_forest.json
similarity index 100%
rename from proposals/ml_fit_regr_random_forest.json
rename to older-proposals/ml_fit_regr_random_forest.json
diff --git a/older-proposals/ml_predict.json b/older-proposals/ml_predict.json
new file mode 100644
index 00000000..87cd2500
--- /dev/null
+++ b/older-proposals/ml_predict.json
@@ -0,0 +1,49 @@
+{
+    "id": "ml_predict",
+    "summary": "Predict using ML",
+    "description": "Applies a machine learning model to a data cube of input features and returns the predicted values.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The data cube containing the input features.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "model",
+            "description": "A ML model that was trained with one of the ML training processes such as ``ml_fit_regr_random_forest()``.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "dimensions",
+            "description": "Zero or more dimensions that will be reduced by the model. Fails with a `DimensionNotAvailable` exception if one of the specified dimensions does not exist.",
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "string"
+                }
+            }
+        }
+    ],
+    "returns": {
+        "description": "A data cube with the predicted values. It removes the specified dimensions and adds new dimension for the predicted values. It has the name `predictions` and is of type `other`. If a single value is returned, the dimension has a single label with name `0`.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube",
+            "dimensions": [
+                {
+                    "type": "other"
+                }
+            ]
+        }
+    }
+}
diff --git a/proposals/nan.json b/older-proposals/nan.json
similarity index 100%
rename from proposals/nan.json
rename to older-proposals/nan.json
diff --git a/proposals/predict_curve.json b/older-proposals/predict_curve.json
similarity index 100%
rename from proposals/predict_curve.json
rename to older-proposals/predict_curve.json
diff --git a/proposals/reduce_spatial.json b/older-proposals/reduce_spatial.json
similarity index 100%
rename from proposals/reduce_spatial.json
rename to older-proposals/reduce_spatial.json
diff --git a/proposals/run_udf_externally.json b/older-proposals/run_udf_externally.json
similarity index 100%
rename from proposals/run_udf_externally.json
rename to older-proposals/run_udf_externally.json
diff --git a/proposals/sar_backscatter.json b/older-proposals/sar_backscatter.json
similarity index 100%
rename from proposals/sar_backscatter.json
rename to older-proposals/sar_backscatter.json
diff --git a/older-proposals/save_ml_model.json b/older-proposals/save_ml_model.json
new file mode 100644
index 00000000..5e9ea8b0
--- /dev/null
+++ b/older-proposals/save_ml_model.json
@@ -0,0 +1,44 @@
+{
+    "id": "save_ml_model",
+    "summary": "Save a ML model",
+    "description": "Saves a machine learning model as part of a batch job.\n\nThe model will be accompanied by a separate STAC Item that implements the [ml-model extension](https://github.com/stac-extensions/ml-model).",
+    "categories": [
+        "machine learning",
+        "import"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The data to store as a machine learning model.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "options",
+            "description": "Additional parameters to create the file(s).",
+            "schema": {
+                "type": "object",
+                "additionalParameters": false
+            },
+            "default": {},
+            "optional": true
+        }
+    ],
+    "returns": {
+        "description": "Returns `false` if the process failed to store the model, `true` otherwise.",
+        "schema": {
+            "type": "boolean"
+        }
+    },
+    "links": [
+        {
+            "href": "https://github.com/stac-extensions/ml-model",
+            "title": "STAC ml-model extension",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
\ No newline at end of file
diff --git a/proposals/unflatten_dimension.json b/older-proposals/unflatten_dimension.json
similarity index 100%
rename from proposals/unflatten_dimension.json
rename to older-proposals/unflatten_dimension.json
diff --git a/proposals/vector_buffer.json b/older-proposals/vector_buffer.json
similarity index 100%
rename from proposals/vector_buffer.json
rename to older-proposals/vector_buffer.json
diff --git a/proposals/vector_reproject.json b/older-proposals/vector_reproject.json
similarity index 100%
rename from proposals/vector_reproject.json
rename to older-proposals/vector_reproject.json
diff --git a/proposals/vector_to_random_points.json b/older-proposals/vector_to_random_points.json
similarity index 100%
rename from proposals/vector_to_random_points.json
rename to older-proposals/vector_to_random_points.json
diff --git a/proposals/vector_to_regular_points.json b/older-proposals/vector_to_regular_points.json
similarity index 100%
rename from proposals/vector_to_regular_points.json
rename to older-proposals/vector_to_regular_points.json
diff --git a/proposals/ML-README.md b/proposals/ML-README.md
new file mode 100644
index 00000000..8fad28fc
--- /dev/null
+++ b/proposals/ML-README.md
@@ -0,0 +1,159 @@
+# ML API Specification for Earth Observation Data Cubes
+
+Welcome to the Machine Learning (ML) API for processing Earth Observation (EO) data cubes. This API provides capabilities to perform classification, regression, smoothing, uncertainty analysis, and more on EO data. Below is a comprehensive guide to the API's operations.
+
+
+
+## API Operations
+
+Here's a detailed breakdown of the ML processes available in this API.
+
+### General ML Processes
+
+- `mlm_class_*` and `mlm_regr_*`
+  - **Purpose**: Placeholder processes for classification and regression model definition.
+  - **Usage**: These processes encapsulate various machine learning model .
+
+- `ml_fit(model, training_set, target = "label")`
+  - **Purpose**: Train an ML model using the specified training dataset.
+  - **Parameters**:
+    - `model`: The machine learning algorithm/model to be trained.
+    - `training_set`: The dataset used for training.
+    - `target`: The label or target variable in the dataset.
+
+- `ml_predict(data, model)`
+  - **Purpose**: Predict outcomes using a trained model on a given dataset.
+  - **Parameters**:
+    - `data`: The dataset for prediction.
+    - `model`: The trained model to use for prediction.
+
+- `ml_predict_probability(data, model)`
+  - **Purpose**: Similar to `ml_predict`, but returns the probability of each prediction.
+  - **Parameters**:
+    - `data`: The dataset for probabilistic prediction.
+    - `model`: The trained model to use for prediction.
+
+- `ml_smooth_class(data, window_size, neighborhood_fraction)`
+  - **Purpose**: Smooth classification results with spatial filtering.
+  - **Parameters**:
+    - `data`: The dataset to be smoothed.
+    - `window_size`: The size of the moving window.
+    - `neighborhood_fraction`: The fraction of neighbors to consider.
+
+- `ml_uncertainty_class(data, method = "margin")`
+  - **Purpose**: Analyze and assess uncertainty in classification results.
+  - **Parameters**:
+    - `data`: The dataset for uncertainty assessment.
+    - `method`: The method for uncertainty evaluation. Default is "margin".
+
+### Model Management
+
+- `save_ml_model(data, name, tasks = list(), options = {})`
+  - **Purpose**: Save a trained model to persistent storage.
+  - **Parameters**:
+    - `data`: The model data to be saved.
+    - `name`: The identifier for the model.
+    - `tasks`: (Optional) A list of tasks associated with the model.
+    - `options`: (Optional) Additional options for saving the model.
+
+- `load_ml_model(uri)`
+  - **Purpose**: Load a previously saved model from storage.
+  - **Parameters**:
+    - `uri`: The location of the stored model.
+
+- `export_model(model, name, folder)`
+  - **Purpose**: Export a model to a specified folder in user workspace.
+  - **Parameters**:
+    - `model`: The model to export.
+    - `name`: The name to save the model as.
+    - `folder`: The destination folder.
+
+- `import_model(name, folder)`
+  - **Purpose**: Import a model from a specified folder in user workspace.
+  - **Parameters**:
+    - `name`: The name of the model to import.
+    - `folder`: The source folder.
+
+### Data Management
+
+- `export_cube(data, name, folder)`
+  - **Purpose**: Export data cubes to a specified folder in user workspace.
+  - **Parameters**:
+    - `data`: The data cube to export.
+    - `name`: The name to save the data cube under.
+    - `folder`: The destination folder.
+
+- `import_cube(name, folder)`
+  - **Purpose**: Import data cubes from a specified folder in user workspace.
+  - **Parameters**:
+    - `name`: The name of the data cube to import.
+    - `folder`: The source folder.
+
+### Results and Validation
+
+- `load_result(job_id)`
+  - **Purpose**: Load the result of a specified job.
+  - **Parameters**:
+    - `job_id`: The identifier for the job whose result is to be loaded.
+
+- `ml_label_class`
+  - **Purpose**: Label the classes in the modeling context.
+  - **Notes**: Use this function to assign labels to data classes post-prediction.
+
+- `ml_validate`
+  - **Purpose**: Validate models and predictions against the dataset.
+  - **Notes**: This function checks for accuracy and reliability of model outputs.
+
+## Examples
+
+TO DO
+
+
+## Naming Convention Explanation
+
+### `mlm_class_` and `mlm_regr_` Prefixes
+
+- **Model Definition**:
+  - The `mlm_class_` prefix is used for functions that define and initialize classification models. These functions create model structures but do not perform any training, returning untrained classification models.
+  - The `mlm_regr_` prefix applies to regression models. These functions provide the framework of a regression model without training, outputting untrained models ready for fitting with data.
+  - This setup allows users to configure the model architecture and parameters prior to training.
+
+### `ml_` Prefix
+
+- **General ML Operations**:
+  - The `ml_` prefix encompasses general machine learning operations applicable across stages, such as model training, prediction, and post-processing. These processes include key tasks like fitting (`ml_fit`), predicting (`ml_predict`), and validating (`ml_validate`).
+
+## Adopting openEO Conventions
+
+### `load_` Prefix
+
+- **Loading Resources**:
+  - The `load_` prefix is used for functions involved in retrieving or loading resources, conforming to openEO's approach to data acquisition.
+  - Examples:
+    - `load_ml_model(uri)`: Loads a machine learning model from storage using its URI.
+    - `load_result(job_id)`: Retrieves the result of a job given its ID.
+
+### `export_` Prefix
+
+- **Exporting Data/Models**:
+  - The `export_` prefix denotes functions that handle exporting data or models.
+  - Examples:
+    - `export_cube(data, name, folder)`: Exports a data cube to a designated folder.
+    - `export_model(model, name, folder)`: Exports a trained model for external use or storage.
+
+### `import_` Prefix
+
+- **Importing Resources**:
+  - The `import_` prefix indicates functions intended to bring data or models from external resources into the application.
+  - Examples:
+    - `import_cube(name, folder)`: Imports a data cube from a designated folder.
+    - `import_model(name, folder)`: Imports a model from a specified folder.
+
+## Advantages of Alignment with openEO
+
+- **Familiarity**: Leveraging openEO conventions helps users quickly understand and operate the API, especially those familiar with openEO.
+- **Consistency**: This reduces confusion by maintaining a uniform interaction model for data, models, and results.
+- **Integration**: Adhering to openEO standards facilitates easier integration with other tools and services within its ecosystem.
+
+This structured approach allows users to effectively transition from defining their models to executing a comprehensive ML workflow tailored for Earth Observation data cubes.
+		
\ No newline at end of file
diff --git a/proposals/load_ml_model.json b/proposals/load_ml_model.json
index 7fa86d89..c2572dac 100644
--- a/proposals/load_ml_model.json
+++ b/proposals/load_ml_model.json
@@ -10,7 +10,7 @@
     "parameters": [
         {
             "name": "uri",
-            "description": "The STAC Item to load the machine learning model from. The STAC Item must implement the `ml-model` extension.",
+            "description": "The STAC Item to load the machine learning model from. The STAC Item must implement the `mlm-model` extension.",
             "schema": [
                 {
                     "title": "URL",
@@ -37,10 +37,10 @@
     },
     "links": [
         {
-            "href": "https://github.com/stac-extensions/ml-model",
-            "title": "STAC ml-model extension",
+            "href": "https://github.com/stac-extensions/mlm",
+            "title": "Machine Learning Model STAC extension",
             "type": "text/html",
             "rel": "about"
         }
     ]
-}
+}
\ No newline at end of file
diff --git a/proposals/ml_fit.json b/proposals/ml_fit.json
new file mode 100644
index 00000000..e2541ed0
--- /dev/null
+++ b/proposals/ml_fit.json
@@ -0,0 +1,63 @@
+{
+    "id": "ml_fit",
+    "summary": "Train a machine learning model",
+    "description": "Executes the fit of a specified machine learning model based on training data. \n\nThe function is generic and supports different machine learning models.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "model",
+            "description": "The machine learning model to be trained. This should be an instance of a model that supports the `ml_fit` method.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "training_set",
+            "description": "The training set for the model, provided as a vector data cube. This set contains both the independent variables and the dependent variable that the model analyzes to learn patterns and relationships within the data.",
+            "schema": [
+                {
+                    "type": "object",
+                    "subtype": "datacube",
+                    "dimensions": [
+                        {
+                            "type": "geometry"
+                        },
+                        {
+                            "type": "bands"
+                        }
+                    ]
+                },
+                {
+                    "type": "object",
+                    "subtype": "datacube",
+                    "dimensions": [
+                        {
+                            "type": "geometry"
+                        },
+                        {
+                            "type": "other"
+                        }
+                    ]
+                }
+            ]
+        },
+        {
+            "name": "target",
+            "description": "The column in the training set that represents the dependent variable for model training.",
+            "schema": {
+                "type": "string"
+            }
+        }
+    ],
+    "returns": {
+        "description": "A trained model object that can be saved with `save_ml_model()` and restored with `load_ml_model()`.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    }
+}
diff --git a/proposals/ml_predict.json b/proposals/ml_predict.json
index 87cd2500..6c606a60 100644
--- a/proposals/ml_predict.json
+++ b/proposals/ml_predict.json
@@ -17,10 +17,10 @@
         },
         {
             "name": "model",
-            "description": "A ML model that was trained with one of the ML training processes such as ``ml_fit_regr_random_forest()``.",
+            "description": "A ML model that was trained.",
             "schema": {
                 "type": "object",
-                "subtype": "ml-model"
+                "subtype": "mlm-model"
             }
         },
         {
@@ -46,4 +46,4 @@
             ]
         }
     }
-}
+}
\ No newline at end of file
diff --git a/proposals/mlm_class_random_forest.json b/proposals/mlm_class_random_forest.json
new file mode 100644
index 00000000..3c4d5b4c
--- /dev/null
+++ b/proposals/mlm_class_random_forest.json
@@ -0,0 +1,67 @@
+{
+    "id": "mlm_class_random_forest",
+    "summary": "Initialize a random forest classification model for later training",
+    "description": "Initializes a Random Forest classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "max_variables",
+            "description": "Specifies how many split variables will be used at a node.\n\nThe following options are available:\n\n- *integer*: The given number of variables are considered for each split.\n- `all`: All variables are considered for each split.\n- `log2`: The logarithm with base 2 of the number of variables are considered for each split.\n- `onethird`: A third of the number of variables are considered for each split.\n- `sqrt`: The square root of the number of variables are considered for each split. This is often the default for classification.",
+            "schema": [
+                {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                {
+                    "type": "string",
+                    "enum": [
+                        "all",
+                        "log2",
+                        "onethird",
+                        "sqrt"
+                    ]
+                }
+            ]
+        },
+        {
+            "name": "num_trees",
+            "description": "The number of trees build within the Random Forest classification.",
+            "optional": true,
+            "default": 100,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for the random sampling in training. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.1023/A:1010933404324",
+            "title": "Breiman (2001): Random Forests",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
\ No newline at end of file
diff --git a/proposals/mlm_regr_random_forest.json b/proposals/mlm_regr_random_forest.json
new file mode 100644
index 00000000..1cdacf47
--- /dev/null
+++ b/proposals/mlm_regr_random_forest.json
@@ -0,0 +1,67 @@
+{
+    "id": "mlm_regr_random_forest",
+    "summary": "Initialize a random forest regression model for later training",
+    "description": "Initializes a Random Forest regression model. This component prepares the model structure but does not perform training. The model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "max_variables",
+            "description": "Specifies how many split variables will be used at a node.\n\nThe following options are available:\n\n- *integer*: The given number of variables are considered for each split.\n- `all`: All variables are considered for each split.\n- `log2`: The logarithm with base 2 of the number of variables are considered for each split.\n- `onethird`: A third of the number of variables are considered for each split. This is often the default for regression.\n- `sqrt`: The square root of the number of variables are considered for each split.",
+            "schema": [
+                {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                {
+                    "type": "string",
+                    "enum": [
+                        "all",
+                        "log2",
+                        "onethird",
+                        "sqrt"
+                    ]
+                }
+            ]
+        },
+        {
+            "name": "num_trees",
+            "description": "The number of trees build within the Random Forest regression.",
+            "optional": true,
+            "default": 100,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for the random sampling in training. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "mlm-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.1023/A:1010933404324",
+            "title": "Breiman (2001): Random Forests",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
\ No newline at end of file
diff --git a/proposals/save_ml_model.json b/proposals/save_ml_model.json
index 5e9ea8b0..a9622610 100644
--- a/proposals/save_ml_model.json
+++ b/proposals/save_ml_model.json
@@ -1,7 +1,7 @@
 {
     "id": "save_ml_model",
     "summary": "Save a ML model",
-    "description": "Saves a machine learning model as part of a batch job.\n\nThe model will be accompanied by a separate STAC Item that implements the [ml-model extension](https://github.com/stac-extensions/ml-model).",
+    "description": "Saves a machine learning model as part of a batch job.\n\nThe model will be accompanied by a separate STAC Item that implements the [mlm-model extension](https://github.com/stac-extensions/mlm).",
     "categories": [
         "machine learning",
         "import"
@@ -35,8 +35,8 @@
     },
     "links": [
         {
-            "href": "https://github.com/stac-extensions/ml-model",
-            "title": "STAC ml-model extension",
+            "href": "https://github.com/stac-extensions/mlm",
+            "title": "Machine Learning Model STAC extension",
             "type": "text/html",
             "rel": "about"
         }

From abf1ca162f8166d8aacc777446e4b54e0e858490 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Tue, 8 Apr 2025 19:40:36 +0200
Subject: [PATCH 05/52] mlm word

---
 .../aggregate_spatial_window.json                                | 0
 {older-proposals => proposals-reserve}/apply_polygon.json        | 0
 .../ard_normalized_radar_backscatter.json                        | 0
 .../ard_surface_reflectance.json                                 | 0
 {older-proposals => proposals-reserve}/array_create_labeled.json | 0
 {older-proposals => proposals-reserve}/array_find_label.json     | 0
 {older-proposals => proposals-reserve}/array_modify.json         | 0
 .../atmospheric_correction.json                                  | 0
 {older-proposals => proposals-reserve}/cloud_detection.json      | 0
 {older-proposals => proposals-reserve}/cummax.json               | 0
 {older-proposals => proposals-reserve}/cummin.json               | 0
 {older-proposals => proposals-reserve}/cumproduct.json           | 0
 {older-proposals => proposals-reserve}/cumsum.json               | 0
 {older-proposals => proposals-reserve}/date_between.json         | 0
 {older-proposals => proposals-reserve}/date_difference.json      | 0
 {older-proposals => proposals-reserve}/date_shift.json           | 0
 {older-proposals => proposals-reserve}/filter_labels.json        | 0
 {older-proposals => proposals-reserve}/filter_vector.json        | 0
 {older-proposals => proposals-reserve}/fit_curve.json            | 0
 {older-proposals => proposals-reserve}/flatten_dimensions.json   | 0
 {older-proposals => proposals-reserve}/inspect.json              | 0
 {older-proposals => proposals-reserve}/is_infinite.json          | 0
 {older-proposals => proposals-reserve}/load_geojson.json         | 0
 {older-proposals => proposals-reserve}/load_ml_model.json        | 0
 {older-proposals => proposals-reserve}/load_stac.json            | 0
 {older-proposals => proposals-reserve}/load_uploaded_files.json  | 0
 {older-proposals => proposals-reserve}/load_url.json             | 0
 .../ml_fit_class_random_forest.json                              | 0
 .../ml_fit_regr_random_forest.json                               | 0
 {older-proposals => proposals-reserve}/ml_predict.json           | 0
 {older-proposals => proposals-reserve}/nan.json                  | 0
 {older-proposals => proposals-reserve}/predict_curve.json        | 0
 {older-proposals => proposals-reserve}/reduce_spatial.json       | 0
 {older-proposals => proposals-reserve}/run_udf_externally.json   | 0
 {older-proposals => proposals-reserve}/sar_backscatter.json      | 0
 {older-proposals => proposals-reserve}/save_ml_model.json        | 0
 {older-proposals => proposals-reserve}/unflatten_dimension.json  | 0
 {older-proposals => proposals-reserve}/vector_buffer.json        | 0
 {older-proposals => proposals-reserve}/vector_reproject.json     | 0
 .../vector_to_random_points.json                                 | 0
 .../vector_to_regular_points.json                                | 0
 tests/.words                                                     | 1 +
 42 files changed, 1 insertion(+)
 rename {older-proposals => proposals-reserve}/aggregate_spatial_window.json (100%)
 rename {older-proposals => proposals-reserve}/apply_polygon.json (100%)
 rename {older-proposals => proposals-reserve}/ard_normalized_radar_backscatter.json (100%)
 rename {older-proposals => proposals-reserve}/ard_surface_reflectance.json (100%)
 rename {older-proposals => proposals-reserve}/array_create_labeled.json (100%)
 rename {older-proposals => proposals-reserve}/array_find_label.json (100%)
 rename {older-proposals => proposals-reserve}/array_modify.json (100%)
 rename {older-proposals => proposals-reserve}/atmospheric_correction.json (100%)
 rename {older-proposals => proposals-reserve}/cloud_detection.json (100%)
 rename {older-proposals => proposals-reserve}/cummax.json (100%)
 rename {older-proposals => proposals-reserve}/cummin.json (100%)
 rename {older-proposals => proposals-reserve}/cumproduct.json (100%)
 rename {older-proposals => proposals-reserve}/cumsum.json (100%)
 rename {older-proposals => proposals-reserve}/date_between.json (100%)
 rename {older-proposals => proposals-reserve}/date_difference.json (100%)
 rename {older-proposals => proposals-reserve}/date_shift.json (100%)
 rename {older-proposals => proposals-reserve}/filter_labels.json (100%)
 rename {older-proposals => proposals-reserve}/filter_vector.json (100%)
 rename {older-proposals => proposals-reserve}/fit_curve.json (100%)
 rename {older-proposals => proposals-reserve}/flatten_dimensions.json (100%)
 rename {older-proposals => proposals-reserve}/inspect.json (100%)
 rename {older-proposals => proposals-reserve}/is_infinite.json (100%)
 rename {older-proposals => proposals-reserve}/load_geojson.json (100%)
 rename {older-proposals => proposals-reserve}/load_ml_model.json (100%)
 rename {older-proposals => proposals-reserve}/load_stac.json (100%)
 rename {older-proposals => proposals-reserve}/load_uploaded_files.json (100%)
 rename {older-proposals => proposals-reserve}/load_url.json (100%)
 rename {older-proposals => proposals-reserve}/ml_fit_class_random_forest.json (100%)
 rename {older-proposals => proposals-reserve}/ml_fit_regr_random_forest.json (100%)
 rename {older-proposals => proposals-reserve}/ml_predict.json (100%)
 rename {older-proposals => proposals-reserve}/nan.json (100%)
 rename {older-proposals => proposals-reserve}/predict_curve.json (100%)
 rename {older-proposals => proposals-reserve}/reduce_spatial.json (100%)
 rename {older-proposals => proposals-reserve}/run_udf_externally.json (100%)
 rename {older-proposals => proposals-reserve}/sar_backscatter.json (100%)
 rename {older-proposals => proposals-reserve}/save_ml_model.json (100%)
 rename {older-proposals => proposals-reserve}/unflatten_dimension.json (100%)
 rename {older-proposals => proposals-reserve}/vector_buffer.json (100%)
 rename {older-proposals => proposals-reserve}/vector_reproject.json (100%)
 rename {older-proposals => proposals-reserve}/vector_to_random_points.json (100%)
 rename {older-proposals => proposals-reserve}/vector_to_regular_points.json (100%)

diff --git a/older-proposals/aggregate_spatial_window.json b/proposals-reserve/aggregate_spatial_window.json
similarity index 100%
rename from older-proposals/aggregate_spatial_window.json
rename to proposals-reserve/aggregate_spatial_window.json
diff --git a/older-proposals/apply_polygon.json b/proposals-reserve/apply_polygon.json
similarity index 100%
rename from older-proposals/apply_polygon.json
rename to proposals-reserve/apply_polygon.json
diff --git a/older-proposals/ard_normalized_radar_backscatter.json b/proposals-reserve/ard_normalized_radar_backscatter.json
similarity index 100%
rename from older-proposals/ard_normalized_radar_backscatter.json
rename to proposals-reserve/ard_normalized_radar_backscatter.json
diff --git a/older-proposals/ard_surface_reflectance.json b/proposals-reserve/ard_surface_reflectance.json
similarity index 100%
rename from older-proposals/ard_surface_reflectance.json
rename to proposals-reserve/ard_surface_reflectance.json
diff --git a/older-proposals/array_create_labeled.json b/proposals-reserve/array_create_labeled.json
similarity index 100%
rename from older-proposals/array_create_labeled.json
rename to proposals-reserve/array_create_labeled.json
diff --git a/older-proposals/array_find_label.json b/proposals-reserve/array_find_label.json
similarity index 100%
rename from older-proposals/array_find_label.json
rename to proposals-reserve/array_find_label.json
diff --git a/older-proposals/array_modify.json b/proposals-reserve/array_modify.json
similarity index 100%
rename from older-proposals/array_modify.json
rename to proposals-reserve/array_modify.json
diff --git a/older-proposals/atmospheric_correction.json b/proposals-reserve/atmospheric_correction.json
similarity index 100%
rename from older-proposals/atmospheric_correction.json
rename to proposals-reserve/atmospheric_correction.json
diff --git a/older-proposals/cloud_detection.json b/proposals-reserve/cloud_detection.json
similarity index 100%
rename from older-proposals/cloud_detection.json
rename to proposals-reserve/cloud_detection.json
diff --git a/older-proposals/cummax.json b/proposals-reserve/cummax.json
similarity index 100%
rename from older-proposals/cummax.json
rename to proposals-reserve/cummax.json
diff --git a/older-proposals/cummin.json b/proposals-reserve/cummin.json
similarity index 100%
rename from older-proposals/cummin.json
rename to proposals-reserve/cummin.json
diff --git a/older-proposals/cumproduct.json b/proposals-reserve/cumproduct.json
similarity index 100%
rename from older-proposals/cumproduct.json
rename to proposals-reserve/cumproduct.json
diff --git a/older-proposals/cumsum.json b/proposals-reserve/cumsum.json
similarity index 100%
rename from older-proposals/cumsum.json
rename to proposals-reserve/cumsum.json
diff --git a/older-proposals/date_between.json b/proposals-reserve/date_between.json
similarity index 100%
rename from older-proposals/date_between.json
rename to proposals-reserve/date_between.json
diff --git a/older-proposals/date_difference.json b/proposals-reserve/date_difference.json
similarity index 100%
rename from older-proposals/date_difference.json
rename to proposals-reserve/date_difference.json
diff --git a/older-proposals/date_shift.json b/proposals-reserve/date_shift.json
similarity index 100%
rename from older-proposals/date_shift.json
rename to proposals-reserve/date_shift.json
diff --git a/older-proposals/filter_labels.json b/proposals-reserve/filter_labels.json
similarity index 100%
rename from older-proposals/filter_labels.json
rename to proposals-reserve/filter_labels.json
diff --git a/older-proposals/filter_vector.json b/proposals-reserve/filter_vector.json
similarity index 100%
rename from older-proposals/filter_vector.json
rename to proposals-reserve/filter_vector.json
diff --git a/older-proposals/fit_curve.json b/proposals-reserve/fit_curve.json
similarity index 100%
rename from older-proposals/fit_curve.json
rename to proposals-reserve/fit_curve.json
diff --git a/older-proposals/flatten_dimensions.json b/proposals-reserve/flatten_dimensions.json
similarity index 100%
rename from older-proposals/flatten_dimensions.json
rename to proposals-reserve/flatten_dimensions.json
diff --git a/older-proposals/inspect.json b/proposals-reserve/inspect.json
similarity index 100%
rename from older-proposals/inspect.json
rename to proposals-reserve/inspect.json
diff --git a/older-proposals/is_infinite.json b/proposals-reserve/is_infinite.json
similarity index 100%
rename from older-proposals/is_infinite.json
rename to proposals-reserve/is_infinite.json
diff --git a/older-proposals/load_geojson.json b/proposals-reserve/load_geojson.json
similarity index 100%
rename from older-proposals/load_geojson.json
rename to proposals-reserve/load_geojson.json
diff --git a/older-proposals/load_ml_model.json b/proposals-reserve/load_ml_model.json
similarity index 100%
rename from older-proposals/load_ml_model.json
rename to proposals-reserve/load_ml_model.json
diff --git a/older-proposals/load_stac.json b/proposals-reserve/load_stac.json
similarity index 100%
rename from older-proposals/load_stac.json
rename to proposals-reserve/load_stac.json
diff --git a/older-proposals/load_uploaded_files.json b/proposals-reserve/load_uploaded_files.json
similarity index 100%
rename from older-proposals/load_uploaded_files.json
rename to proposals-reserve/load_uploaded_files.json
diff --git a/older-proposals/load_url.json b/proposals-reserve/load_url.json
similarity index 100%
rename from older-proposals/load_url.json
rename to proposals-reserve/load_url.json
diff --git a/older-proposals/ml_fit_class_random_forest.json b/proposals-reserve/ml_fit_class_random_forest.json
similarity index 100%
rename from older-proposals/ml_fit_class_random_forest.json
rename to proposals-reserve/ml_fit_class_random_forest.json
diff --git a/older-proposals/ml_fit_regr_random_forest.json b/proposals-reserve/ml_fit_regr_random_forest.json
similarity index 100%
rename from older-proposals/ml_fit_regr_random_forest.json
rename to proposals-reserve/ml_fit_regr_random_forest.json
diff --git a/older-proposals/ml_predict.json b/proposals-reserve/ml_predict.json
similarity index 100%
rename from older-proposals/ml_predict.json
rename to proposals-reserve/ml_predict.json
diff --git a/older-proposals/nan.json b/proposals-reserve/nan.json
similarity index 100%
rename from older-proposals/nan.json
rename to proposals-reserve/nan.json
diff --git a/older-proposals/predict_curve.json b/proposals-reserve/predict_curve.json
similarity index 100%
rename from older-proposals/predict_curve.json
rename to proposals-reserve/predict_curve.json
diff --git a/older-proposals/reduce_spatial.json b/proposals-reserve/reduce_spatial.json
similarity index 100%
rename from older-proposals/reduce_spatial.json
rename to proposals-reserve/reduce_spatial.json
diff --git a/older-proposals/run_udf_externally.json b/proposals-reserve/run_udf_externally.json
similarity index 100%
rename from older-proposals/run_udf_externally.json
rename to proposals-reserve/run_udf_externally.json
diff --git a/older-proposals/sar_backscatter.json b/proposals-reserve/sar_backscatter.json
similarity index 100%
rename from older-proposals/sar_backscatter.json
rename to proposals-reserve/sar_backscatter.json
diff --git a/older-proposals/save_ml_model.json b/proposals-reserve/save_ml_model.json
similarity index 100%
rename from older-proposals/save_ml_model.json
rename to proposals-reserve/save_ml_model.json
diff --git a/older-proposals/unflatten_dimension.json b/proposals-reserve/unflatten_dimension.json
similarity index 100%
rename from older-proposals/unflatten_dimension.json
rename to proposals-reserve/unflatten_dimension.json
diff --git a/older-proposals/vector_buffer.json b/proposals-reserve/vector_buffer.json
similarity index 100%
rename from older-proposals/vector_buffer.json
rename to proposals-reserve/vector_buffer.json
diff --git a/older-proposals/vector_reproject.json b/proposals-reserve/vector_reproject.json
similarity index 100%
rename from older-proposals/vector_reproject.json
rename to proposals-reserve/vector_reproject.json
diff --git a/older-proposals/vector_to_random_points.json b/proposals-reserve/vector_to_random_points.json
similarity index 100%
rename from older-proposals/vector_to_random_points.json
rename to proposals-reserve/vector_to_random_points.json
diff --git a/older-proposals/vector_to_regular_points.json b/proposals-reserve/vector_to_regular_points.json
similarity index 100%
rename from older-proposals/vector_to_regular_points.json
rename to proposals-reserve/vector_to_regular_points.json
diff --git a/tests/.words b/tests/.words
index a50285ba..86c5e340 100644
--- a/tests/.words
+++ b/tests/.words
@@ -47,3 +47,4 @@ Hyndman
 date1
 date2
 favor
+mlm-model

From 6f6d903a83beb477f60790ca92a5b1d1138ba796 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Tue, 8 Apr 2025 19:49:13 +0200
Subject: [PATCH 06/52] refactor ml rf

---
 proposals/mlm_class_random_forest.json | 2 +-
 proposals/mlm_regr_random_forest.json  | 4 ++--
 tests/.words                           | 1 +
 3 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/proposals/mlm_class_random_forest.json b/proposals/mlm_class_random_forest.json
index 3c4d5b4c..0881e0f0 100644
--- a/proposals/mlm_class_random_forest.json
+++ b/proposals/mlm_class_random_forest.json
@@ -1,6 +1,6 @@
 {
     "id": "mlm_class_random_forest",
-    "summary": "Initialize a random forest classification model for later training",
+    "summary": "Initialize a random forest classification model",
     "description": "Initializes a Random Forest classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
     "categories": [
         "machine learning"
diff --git a/proposals/mlm_regr_random_forest.json b/proposals/mlm_regr_random_forest.json
index 1cdacf47..1eecedc3 100644
--- a/proposals/mlm_regr_random_forest.json
+++ b/proposals/mlm_regr_random_forest.json
@@ -1,6 +1,6 @@
 {
     "id": "mlm_regr_random_forest",
-    "summary": "Initialize a random forest regression model for later training",
+    "summary": "Initialize a random forest regression model",
     "description": "Initializes a Random Forest regression model. This component prepares the model structure but does not perform training. The model can be trained later using ``ml_fit``.",
     "categories": [
         "machine learning"
@@ -53,7 +53,7 @@
         "description": "A model object that can be trained using ``ml_fit``.",
         "schema": {
             "type": "object",
-            "subtype": "mlm-model"
+            "subtype": "ml-model"
         }
     },
     "links": [
diff --git a/tests/.words b/tests/.words
index 86c5e340..2d4e623a 100644
--- a/tests/.words
+++ b/tests/.words
@@ -48,3 +48,4 @@ date1
 date2
 favor
 mlm-model
+analyzes

From 74ae56e716ac6dac0f48d8a8bb248db28e165ff5 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Wed, 9 Apr 2025 10:50:12 +0200
Subject: [PATCH 07/52] refactor files

---
 CHANGELOG.md                                  |   5 +-
 {proposals => proposals-ml}/ML-README.md      |   0
 .../load_ml_model.json                        |   8 +-
 proposals-ml/ml_fit.json                      |  62 ++++++++++
 .../ml_predict.json                           |   7 +-
 .../mlm_class_random_forest.json              |  53 +--------
 proposals-ml/mlm_regr_random_forest.json      |  57 +++++++++
 .../save_ml_model.json                        |   6 +-
 .../ml_fit_regr_random_forest.json            | 110 ------------------
 .../aggregate_spatial_window.json             |   0
 .../apply_polygon.json                        |   0
 .../ard_normalized_radar_backscatter.json     |   0
 .../ard_surface_reflectance.json              |   0
 .../array_create_labeled.json                 |   0
 .../array_find_label.json                     |   0
 .../array_modify.json                         |   0
 .../atmospheric_correction.json               |   0
 .../cloud_detection.json                      |   0
 {proposals-reserve => proposals}/cummax.json  |   0
 {proposals-reserve => proposals}/cummin.json  |   0
 .../cumproduct.json                           |   0
 {proposals-reserve => proposals}/cumsum.json  |   0
 .../date_between.json                         |   0
 .../date_difference.json                      |   0
 .../date_shift.json                           |   0
 .../filter_labels.json                        |   0
 .../filter_vector.json                        |   0
 .../fit_curve.json                            |   0
 .../flatten_dimensions.json                   |   0
 {proposals-reserve => proposals}/inspect.json |   0
 .../is_infinite.json                          |   0
 .../load_geojson.json                         |   0
 .../load_stac.json                            |   0
 .../load_uploaded_files.json                  |   0
 .../load_url.json                             |   0
 proposals/ml_fit.json                         |   7 +-
 proposals/ml_predict.json                     |   9 +-
 proposals/mlm_regr_random_forest.json         |  18 +--
 {proposals-reserve => proposals}/nan.json     |   0
 .../predict_curve.json                        |   0
 .../reduce_spatial.json                       |   0
 .../run_udf_externally.json                   |   0
 .../sar_backscatter.json                      |   0
 .../unflatten_dimension.json                  |   0
 .../vector_buffer.json                        |   0
 .../vector_reproject.json                     |   0
 .../vector_to_random_points.json              |   0
 .../vector_to_regular_points.json             |   0
 48 files changed, 148 insertions(+), 194 deletions(-)
 rename {proposals => proposals-ml}/ML-README.md (100%)
 rename {proposals-reserve => proposals-ml}/load_ml_model.json (87%)
 create mode 100644 proposals-ml/ml_fit.json
 rename {proposals-reserve => proposals-ml}/ml_predict.json (88%)
 rename proposals-reserve/ml_fit_class_random_forest.json => proposals-ml/mlm_class_random_forest.json (51%)
 create mode 100644 proposals-ml/mlm_regr_random_forest.json
 rename {proposals-reserve => proposals-ml}/save_ml_model.json (84%)
 delete mode 100644 proposals-reserve/ml_fit_regr_random_forest.json
 rename {proposals-reserve => proposals}/aggregate_spatial_window.json (100%)
 rename {proposals-reserve => proposals}/apply_polygon.json (100%)
 rename {proposals-reserve => proposals}/ard_normalized_radar_backscatter.json (100%)
 rename {proposals-reserve => proposals}/ard_surface_reflectance.json (100%)
 rename {proposals-reserve => proposals}/array_create_labeled.json (100%)
 rename {proposals-reserve => proposals}/array_find_label.json (100%)
 rename {proposals-reserve => proposals}/array_modify.json (100%)
 rename {proposals-reserve => proposals}/atmospheric_correction.json (100%)
 rename {proposals-reserve => proposals}/cloud_detection.json (100%)
 rename {proposals-reserve => proposals}/cummax.json (100%)
 rename {proposals-reserve => proposals}/cummin.json (100%)
 rename {proposals-reserve => proposals}/cumproduct.json (100%)
 rename {proposals-reserve => proposals}/cumsum.json (100%)
 rename {proposals-reserve => proposals}/date_between.json (100%)
 rename {proposals-reserve => proposals}/date_difference.json (100%)
 rename {proposals-reserve => proposals}/date_shift.json (100%)
 rename {proposals-reserve => proposals}/filter_labels.json (100%)
 rename {proposals-reserve => proposals}/filter_vector.json (100%)
 rename {proposals-reserve => proposals}/fit_curve.json (100%)
 rename {proposals-reserve => proposals}/flatten_dimensions.json (100%)
 rename {proposals-reserve => proposals}/inspect.json (100%)
 rename {proposals-reserve => proposals}/is_infinite.json (100%)
 rename {proposals-reserve => proposals}/load_geojson.json (100%)
 rename {proposals-reserve => proposals}/load_stac.json (100%)
 rename {proposals-reserve => proposals}/load_uploaded_files.json (100%)
 rename {proposals-reserve => proposals}/load_url.json (100%)
 rename {proposals-reserve => proposals}/nan.json (100%)
 rename {proposals-reserve => proposals}/predict_curve.json (100%)
 rename {proposals-reserve => proposals}/reduce_spatial.json (100%)
 rename {proposals-reserve => proposals}/run_udf_externally.json (100%)
 rename {proposals-reserve => proposals}/sar_backscatter.json (100%)
 rename {proposals-reserve => proposals}/unflatten_dimension.json (100%)
 rename {proposals-reserve => proposals}/vector_buffer.json (100%)
 rename {proposals-reserve => proposals}/vector_reproject.json (100%)
 rename {proposals-reserve => proposals}/vector_to_random_points.json (100%)
 rename {proposals-reserve => proposals}/vector_to_regular_points.json (100%)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index f0b68043..f73ac5c8 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -18,8 +18,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `load_geojson`
     - `load_ml_model`
     - `load_url`
-    - `ml_fit_class_random_forest`
-    - `ml_fit_regr_random_forest`
+    - `mlm_class_random_forest`
+    - `mlm_regr_random_forest`
+    - `ml_fit`
     - `ml_predict`
     - `save_ml_model`
     - `unflatten_dimension`
diff --git a/proposals/ML-README.md b/proposals-ml/ML-README.md
similarity index 100%
rename from proposals/ML-README.md
rename to proposals-ml/ML-README.md
diff --git a/proposals-reserve/load_ml_model.json b/proposals-ml/load_ml_model.json
similarity index 87%
rename from proposals-reserve/load_ml_model.json
rename to proposals-ml/load_ml_model.json
index 7fa86d89..c2572dac 100644
--- a/proposals-reserve/load_ml_model.json
+++ b/proposals-ml/load_ml_model.json
@@ -10,7 +10,7 @@
     "parameters": [
         {
             "name": "uri",
-            "description": "The STAC Item to load the machine learning model from. The STAC Item must implement the `ml-model` extension.",
+            "description": "The STAC Item to load the machine learning model from. The STAC Item must implement the `mlm-model` extension.",
             "schema": [
                 {
                     "title": "URL",
@@ -37,10 +37,10 @@
     },
     "links": [
         {
-            "href": "https://github.com/stac-extensions/ml-model",
-            "title": "STAC ml-model extension",
+            "href": "https://github.com/stac-extensions/mlm",
+            "title": "Machine Learning Model STAC extension",
             "type": "text/html",
             "rel": "about"
         }
     ]
-}
+}
\ No newline at end of file
diff --git a/proposals-ml/ml_fit.json b/proposals-ml/ml_fit.json
new file mode 100644
index 00000000..cd485e8a
--- /dev/null
+++ b/proposals-ml/ml_fit.json
@@ -0,0 +1,62 @@
+{
+    "id": "ml_fit",
+    "summary": "Train a machine learning model",
+    "description": "Executes the fit of a specified machine learning model based on training data.\n\nThe function is generic and supports different machine learning models.",
+    "categories": ["machine learning"],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "model",
+            "description": "The machine learning model to be trained. This should be an instance of a model that supports the `ml_fit` method.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "training_set",
+            "description": "The training set for the model, provided as a vector data cube. This set contains both the independent variables and the dependent variable that the model analyzes to learn patterns and relationships within the data.",
+            "schema": [
+                {
+                    "type": "object",
+                    "subtype": "datacube",
+                    "dimensions": [
+                        {
+                            "type": "geometry"
+                        },
+                        {
+                            "type": "bands"
+                        }
+                    ]
+                },
+                {
+                    "type": "object",
+                    "subtype": "datacube",
+                    "dimensions": [
+                        {
+                            "type": "geometry"
+                        },
+                        {
+                            "type": "other"
+                        }
+                    ]
+                }
+            ]
+        },
+        {
+            "name": "target",
+            "description": "The column in the training set that represents the dependent variable for model training.",
+            "schema": {
+                "type": "string"
+            }
+        }
+    ],
+    "returns": {
+        "description": "A trained model object that can be saved with `save_ml_model()` and restored with `load_ml_model()`.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    }
+}
+
diff --git a/proposals-reserve/ml_predict.json b/proposals-ml/ml_predict.json
similarity index 88%
rename from proposals-reserve/ml_predict.json
rename to proposals-ml/ml_predict.json
index 87cd2500..d090a7a6 100644
--- a/proposals-reserve/ml_predict.json
+++ b/proposals-ml/ml_predict.json
@@ -2,9 +2,7 @@
     "id": "ml_predict",
     "summary": "Predict using ML",
     "description": "Applies a machine learning model to a data cube of input features and returns the predicted values.",
-    "categories": [
-        "machine learning"
-    ],
+    "categories": ["machine learning"],
     "experimental": true,
     "parameters": [
         {
@@ -17,7 +15,7 @@
         },
         {
             "name": "model",
-            "description": "A ML model that was trained with one of the ML training processes such as ``ml_fit_regr_random_forest()``.",
+            "description": "A ML model that was trained.",
             "schema": {
                 "type": "object",
                 "subtype": "ml-model"
@@ -47,3 +45,4 @@
         }
     }
 }
+
diff --git a/proposals-reserve/ml_fit_class_random_forest.json b/proposals-ml/mlm_class_random_forest.json
similarity index 51%
rename from proposals-reserve/ml_fit_class_random_forest.json
rename to proposals-ml/mlm_class_random_forest.json
index 63da48a1..0881e0f0 100644
--- a/proposals-reserve/ml_fit_class_random_forest.json
+++ b/proposals-ml/mlm_class_random_forest.json
@@ -1,55 +1,12 @@
 {
-    "id": "ml_fit_class_random_forest",
-    "summary": "Train a random forest classification model",
-    "description": "Executes the fit of a random forest classification based on training data. The process does not include a separate split of the data in test, validation and training data. The Random Forest classification model is based on the approach by Breiman (2001).",
+    "id": "mlm_class_random_forest",
+    "summary": "Initialize a random forest classification model",
+    "description": "Initializes a Random Forest classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
     "categories": [
         "machine learning"
     ],
     "experimental": true,
     "parameters": [
-        {
-            "name": "predictors",
-            "description": "The predictors for the classification model as a vector data cube. Aggregated to the features (vectors) of the target input variable.",
-            "schema": [
-                {
-                    "type": "object",
-                    "subtype": "datacube",
-                    "dimensions": [
-                        {
-                            "type": "geometry"
-                        },
-                        {
-                            "type": "bands"
-                        }
-                    ]
-                },
-                {
-                    "type": "object",
-                    "subtype": "datacube",
-                    "dimensions": [
-                        {
-                            "type": "geometry"
-                        },
-                        {
-                            "type": "other"
-                        }
-                    ]
-                }
-            ]
-        },
-        {
-            "name": "target",
-            "description": "The training sites for the classification model as a vector data cube. This is associated with the target variable for the Random Forest model. The geometry has to associated with a value to predict (e.g. fractional forest canopy cover).",
-            "schema": {
-                "type": "object",
-                "subtype": "datacube",
-                "dimensions": [
-                    {
-                        "type": "geometry"
-                    }
-                ]
-            }
-        },
         {
             "name": "max_variables",
             "description": "Specifies how many split variables will be used at a node.\n\nThe following options are available:\n\n- *integer*: The given number of variables are considered for each split.\n- `all`: All variables are considered for each split.\n- `log2`: The logarithm with base 2 of the number of variables are considered for each split.\n- `onethird`: A third of the number of variables are considered for each split.\n- `sqrt`: The square root of the number of variables are considered for each split. This is often the default for classification.",
@@ -93,7 +50,7 @@
         }
     ],
     "returns": {
-        "description": "A model object that can be saved with ``save_ml_model()`` and restored with ``load_ml_model()``.",
+        "description": "A model object that can be trained using ``ml_fit``.",
         "schema": {
             "type": "object",
             "subtype": "ml-model"
@@ -107,4 +64,4 @@
             "rel": "about"
         }
     ]
-}
+}
\ No newline at end of file
diff --git a/proposals-ml/mlm_regr_random_forest.json b/proposals-ml/mlm_regr_random_forest.json
new file mode 100644
index 00000000..78688f4a
--- /dev/null
+++ b/proposals-ml/mlm_regr_random_forest.json
@@ -0,0 +1,57 @@
+{
+    "id": "mlm_regr_random_forest",
+    "summary": "Initialize a random forest regression model",
+    "description": "Initializes a Random Forest regression model. This component prepares the model structure but does not perform training. The model can be trained later using ``ml_fit``.",
+    "categories": ["machine learning"],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "max_variables",
+            "description": "Specifies how many split variables will be used at a node.\n\nThe following options are available:\n\n- *integer*: The given number of variables are considered for each split.\n- `all`: All variables are considered for each split.\n- `log2`: The logarithm with base 2 of the number of variables are considered for each split.\n- `onethird`: A third of the number of variables are considered for each split. This is often the default for regression.\n- `sqrt`: The square root of the number of variables are considered for each split.",
+            "schema": [
+                {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                {
+                    "type": "string",
+                    "enum": ["all", "log2", "onethird", "sqrt"]
+                }
+            ]
+        },
+        {
+            "name": "num_trees",
+            "description": "The number of trees build within the Random Forest regression.",
+            "optional": true,
+            "default": 100,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for the random sampling in training. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": ["integer", "null"]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.1023/A:1010933404324",
+            "title": "Breiman (2001): Random Forests",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
diff --git a/proposals-reserve/save_ml_model.json b/proposals-ml/save_ml_model.json
similarity index 84%
rename from proposals-reserve/save_ml_model.json
rename to proposals-ml/save_ml_model.json
index 5e9ea8b0..a9622610 100644
--- a/proposals-reserve/save_ml_model.json
+++ b/proposals-ml/save_ml_model.json
@@ -1,7 +1,7 @@
 {
     "id": "save_ml_model",
     "summary": "Save a ML model",
-    "description": "Saves a machine learning model as part of a batch job.\n\nThe model will be accompanied by a separate STAC Item that implements the [ml-model extension](https://github.com/stac-extensions/ml-model).",
+    "description": "Saves a machine learning model as part of a batch job.\n\nThe model will be accompanied by a separate STAC Item that implements the [mlm-model extension](https://github.com/stac-extensions/mlm).",
     "categories": [
         "machine learning",
         "import"
@@ -35,8 +35,8 @@
     },
     "links": [
         {
-            "href": "https://github.com/stac-extensions/ml-model",
-            "title": "STAC ml-model extension",
+            "href": "https://github.com/stac-extensions/mlm",
+            "title": "Machine Learning Model STAC extension",
             "type": "text/html",
             "rel": "about"
         }
diff --git a/proposals-reserve/ml_fit_regr_random_forest.json b/proposals-reserve/ml_fit_regr_random_forest.json
deleted file mode 100644
index 39207324..00000000
--- a/proposals-reserve/ml_fit_regr_random_forest.json
+++ /dev/null
@@ -1,110 +0,0 @@
-{
-    "id": "ml_fit_regr_random_forest",
-    "summary": "Train a random forest regression model",
-    "description": "Executes the fit of a random forest regression based on training data. The process does not include a separate split of the data in test, validation and training data. The Random Forest regression model is based on the approach by Breiman (2001).",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "predictors",
-            "description": "The predictors for the regression model as a vector data cube. Aggregated to the features (vectors) of the target input variable.",
-            "schema": [
-                {
-                    "type": "object",
-                    "subtype": "datacube",
-                    "dimensions": [
-                        {
-                            "type": "geometry"
-                        },
-                        {
-                            "type": "bands"
-                        }
-                    ]
-                },
-                {
-                    "type": "object",
-                    "subtype": "datacube",
-                    "dimensions": [
-                        {
-                            "type": "geometry"
-                        },
-                        {
-                            "type": "other"
-                        }
-                    ]
-                }
-            ]
-        },
-        {
-            "name": "target",
-            "description": "The training sites for the regression model as a vector data cube. This is associated with the target variable for the Random Forest model. The geometry has to associated with a value to predict (e.g. fractional forest canopy cover).",
-            "schema": {
-                "type": "object",
-                "subtype": "datacube",
-                "dimensions": [
-                    {
-                        "type": "geometry"
-                    }
-                ]
-            }
-        },
-        {
-            "name": "max_variables",
-            "description": "Specifies how many split variables will be used at a node.\n\nThe following options are available:\n\n- *integer*: The given number of variables are considered for each split.\n- `all`: All variables are considered for each split.\n- `log2`: The logarithm with base 2 of the number of variables are considered for each split.\n- `onethird`: A third of the number of variables are considered for each split. This is often the default for regression.\n- `sqrt`: The square root of the number of variables are considered for each split.",
-            "schema": [
-                {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                {
-                    "type": "string",
-                    "enum": [
-                        "all",
-                        "log2",
-                        "onethird",
-                        "sqrt"
-                    ]
-                }
-            ]
-        },
-        {
-            "name": "num_trees",
-            "description": "The number of trees build within the Random Forest regression.",
-            "optional": true,
-            "default": 100,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for the random sampling in training. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be saved with ``save_ml_model()`` and restored with ``load_ml_model()``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://doi.org/10.1023/A:1010933404324",
-            "title": "Breiman (2001): Random Forests",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-}
diff --git a/proposals-reserve/aggregate_spatial_window.json b/proposals/aggregate_spatial_window.json
similarity index 100%
rename from proposals-reserve/aggregate_spatial_window.json
rename to proposals/aggregate_spatial_window.json
diff --git a/proposals-reserve/apply_polygon.json b/proposals/apply_polygon.json
similarity index 100%
rename from proposals-reserve/apply_polygon.json
rename to proposals/apply_polygon.json
diff --git a/proposals-reserve/ard_normalized_radar_backscatter.json b/proposals/ard_normalized_radar_backscatter.json
similarity index 100%
rename from proposals-reserve/ard_normalized_radar_backscatter.json
rename to proposals/ard_normalized_radar_backscatter.json
diff --git a/proposals-reserve/ard_surface_reflectance.json b/proposals/ard_surface_reflectance.json
similarity index 100%
rename from proposals-reserve/ard_surface_reflectance.json
rename to proposals/ard_surface_reflectance.json
diff --git a/proposals-reserve/array_create_labeled.json b/proposals/array_create_labeled.json
similarity index 100%
rename from proposals-reserve/array_create_labeled.json
rename to proposals/array_create_labeled.json
diff --git a/proposals-reserve/array_find_label.json b/proposals/array_find_label.json
similarity index 100%
rename from proposals-reserve/array_find_label.json
rename to proposals/array_find_label.json
diff --git a/proposals-reserve/array_modify.json b/proposals/array_modify.json
similarity index 100%
rename from proposals-reserve/array_modify.json
rename to proposals/array_modify.json
diff --git a/proposals-reserve/atmospheric_correction.json b/proposals/atmospheric_correction.json
similarity index 100%
rename from proposals-reserve/atmospheric_correction.json
rename to proposals/atmospheric_correction.json
diff --git a/proposals-reserve/cloud_detection.json b/proposals/cloud_detection.json
similarity index 100%
rename from proposals-reserve/cloud_detection.json
rename to proposals/cloud_detection.json
diff --git a/proposals-reserve/cummax.json b/proposals/cummax.json
similarity index 100%
rename from proposals-reserve/cummax.json
rename to proposals/cummax.json
diff --git a/proposals-reserve/cummin.json b/proposals/cummin.json
similarity index 100%
rename from proposals-reserve/cummin.json
rename to proposals/cummin.json
diff --git a/proposals-reserve/cumproduct.json b/proposals/cumproduct.json
similarity index 100%
rename from proposals-reserve/cumproduct.json
rename to proposals/cumproduct.json
diff --git a/proposals-reserve/cumsum.json b/proposals/cumsum.json
similarity index 100%
rename from proposals-reserve/cumsum.json
rename to proposals/cumsum.json
diff --git a/proposals-reserve/date_between.json b/proposals/date_between.json
similarity index 100%
rename from proposals-reserve/date_between.json
rename to proposals/date_between.json
diff --git a/proposals-reserve/date_difference.json b/proposals/date_difference.json
similarity index 100%
rename from proposals-reserve/date_difference.json
rename to proposals/date_difference.json
diff --git a/proposals-reserve/date_shift.json b/proposals/date_shift.json
similarity index 100%
rename from proposals-reserve/date_shift.json
rename to proposals/date_shift.json
diff --git a/proposals-reserve/filter_labels.json b/proposals/filter_labels.json
similarity index 100%
rename from proposals-reserve/filter_labels.json
rename to proposals/filter_labels.json
diff --git a/proposals-reserve/filter_vector.json b/proposals/filter_vector.json
similarity index 100%
rename from proposals-reserve/filter_vector.json
rename to proposals/filter_vector.json
diff --git a/proposals-reserve/fit_curve.json b/proposals/fit_curve.json
similarity index 100%
rename from proposals-reserve/fit_curve.json
rename to proposals/fit_curve.json
diff --git a/proposals-reserve/flatten_dimensions.json b/proposals/flatten_dimensions.json
similarity index 100%
rename from proposals-reserve/flatten_dimensions.json
rename to proposals/flatten_dimensions.json
diff --git a/proposals-reserve/inspect.json b/proposals/inspect.json
similarity index 100%
rename from proposals-reserve/inspect.json
rename to proposals/inspect.json
diff --git a/proposals-reserve/is_infinite.json b/proposals/is_infinite.json
similarity index 100%
rename from proposals-reserve/is_infinite.json
rename to proposals/is_infinite.json
diff --git a/proposals-reserve/load_geojson.json b/proposals/load_geojson.json
similarity index 100%
rename from proposals-reserve/load_geojson.json
rename to proposals/load_geojson.json
diff --git a/proposals-reserve/load_stac.json b/proposals/load_stac.json
similarity index 100%
rename from proposals-reserve/load_stac.json
rename to proposals/load_stac.json
diff --git a/proposals-reserve/load_uploaded_files.json b/proposals/load_uploaded_files.json
similarity index 100%
rename from proposals-reserve/load_uploaded_files.json
rename to proposals/load_uploaded_files.json
diff --git a/proposals-reserve/load_url.json b/proposals/load_url.json
similarity index 100%
rename from proposals-reserve/load_url.json
rename to proposals/load_url.json
diff --git a/proposals/ml_fit.json b/proposals/ml_fit.json
index e2541ed0..cd485e8a 100644
--- a/proposals/ml_fit.json
+++ b/proposals/ml_fit.json
@@ -1,10 +1,8 @@
 {
     "id": "ml_fit",
     "summary": "Train a machine learning model",
-    "description": "Executes the fit of a specified machine learning model based on training data. \n\nThe function is generic and supports different machine learning models.",
-    "categories": [
-        "machine learning"
-    ],
+    "description": "Executes the fit of a specified machine learning model based on training data.\n\nThe function is generic and supports different machine learning models.",
+    "categories": ["machine learning"],
     "experimental": true,
     "parameters": [
         {
@@ -61,3 +59,4 @@
         }
     }
 }
+
diff --git a/proposals/ml_predict.json b/proposals/ml_predict.json
index 6c606a60..d090a7a6 100644
--- a/proposals/ml_predict.json
+++ b/proposals/ml_predict.json
@@ -2,9 +2,7 @@
     "id": "ml_predict",
     "summary": "Predict using ML",
     "description": "Applies a machine learning model to a data cube of input features and returns the predicted values.",
-    "categories": [
-        "machine learning"
-    ],
+    "categories": ["machine learning"],
     "experimental": true,
     "parameters": [
         {
@@ -20,7 +18,7 @@
             "description": "A ML model that was trained.",
             "schema": {
                 "type": "object",
-                "subtype": "mlm-model"
+                "subtype": "ml-model"
             }
         },
         {
@@ -46,4 +44,5 @@
             ]
         }
     }
-}
\ No newline at end of file
+}
+
diff --git a/proposals/mlm_regr_random_forest.json b/proposals/mlm_regr_random_forest.json
index 1eecedc3..78688f4a 100644
--- a/proposals/mlm_regr_random_forest.json
+++ b/proposals/mlm_regr_random_forest.json
@@ -2,9 +2,7 @@
     "id": "mlm_regr_random_forest",
     "summary": "Initialize a random forest regression model",
     "description": "Initializes a Random Forest regression model. This component prepares the model structure but does not perform training. The model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
+    "categories": ["machine learning"],
     "experimental": true,
     "parameters": [
         {
@@ -17,12 +15,7 @@
                 },
                 {
                     "type": "string",
-                    "enum": [
-                        "all",
-                        "log2",
-                        "onethird",
-                        "sqrt"
-                    ]
+                    "enum": ["all", "log2", "onethird", "sqrt"]
                 }
             ]
         },
@@ -42,10 +35,7 @@
             "optional": true,
             "default": null,
             "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
+                "type": ["integer", "null"]
             }
         }
     ],
@@ -64,4 +54,4 @@
             "rel": "about"
         }
     ]
-}
\ No newline at end of file
+}
diff --git a/proposals-reserve/nan.json b/proposals/nan.json
similarity index 100%
rename from proposals-reserve/nan.json
rename to proposals/nan.json
diff --git a/proposals-reserve/predict_curve.json b/proposals/predict_curve.json
similarity index 100%
rename from proposals-reserve/predict_curve.json
rename to proposals/predict_curve.json
diff --git a/proposals-reserve/reduce_spatial.json b/proposals/reduce_spatial.json
similarity index 100%
rename from proposals-reserve/reduce_spatial.json
rename to proposals/reduce_spatial.json
diff --git a/proposals-reserve/run_udf_externally.json b/proposals/run_udf_externally.json
similarity index 100%
rename from proposals-reserve/run_udf_externally.json
rename to proposals/run_udf_externally.json
diff --git a/proposals-reserve/sar_backscatter.json b/proposals/sar_backscatter.json
similarity index 100%
rename from proposals-reserve/sar_backscatter.json
rename to proposals/sar_backscatter.json
diff --git a/proposals-reserve/unflatten_dimension.json b/proposals/unflatten_dimension.json
similarity index 100%
rename from proposals-reserve/unflatten_dimension.json
rename to proposals/unflatten_dimension.json
diff --git a/proposals-reserve/vector_buffer.json b/proposals/vector_buffer.json
similarity index 100%
rename from proposals-reserve/vector_buffer.json
rename to proposals/vector_buffer.json
diff --git a/proposals-reserve/vector_reproject.json b/proposals/vector_reproject.json
similarity index 100%
rename from proposals-reserve/vector_reproject.json
rename to proposals/vector_reproject.json
diff --git a/proposals-reserve/vector_to_random_points.json b/proposals/vector_to_random_points.json
similarity index 100%
rename from proposals-reserve/vector_to_random_points.json
rename to proposals/vector_to_random_points.json
diff --git a/proposals-reserve/vector_to_regular_points.json b/proposals/vector_to_regular_points.json
similarity index 100%
rename from proposals-reserve/vector_to_regular_points.json
rename to proposals/vector_to_regular_points.json

From f3365fc6d28126e516a55547ebca4b31f9dbd3c8 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Wed, 9 Apr 2025 11:05:16 +0200
Subject: [PATCH 08/52] ml regr description

---
 proposals-ml/load_ml_model.json          | 10 ++++------
 proposals-ml/mlm_regr_random_forest.json |  1 +
 proposals/load_ml_model.json             | 10 ++++------
 proposals/mlm_regr_random_forest.json    |  1 +
 4 files changed, 10 insertions(+), 12 deletions(-)

diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_ml_model.json
index c2572dac..8a54cf45 100644
--- a/proposals-ml/load_ml_model.json
+++ b/proposals-ml/load_ml_model.json
@@ -1,11 +1,8 @@
 {
     "id": "load_ml_model",
     "summary": "Load a ML model",
-    "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit_regr_random_forest()`` and ``save_ml_model()``.",
-    "categories": [
-        "machine learning",
-        "import"
-    ],
+    "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()``.",
+    "categories": ["machine learning", "import"],
     "experimental": true,
     "parameters": [
         {
@@ -43,4 +40,5 @@
             "rel": "about"
         }
     ]
-}
\ No newline at end of file
+}
+
diff --git a/proposals-ml/mlm_regr_random_forest.json b/proposals-ml/mlm_regr_random_forest.json
index 78688f4a..088e2cb0 100644
--- a/proposals-ml/mlm_regr_random_forest.json
+++ b/proposals-ml/mlm_regr_random_forest.json
@@ -55,3 +55,4 @@
         }
     ]
 }
+
diff --git a/proposals/load_ml_model.json b/proposals/load_ml_model.json
index c2572dac..8a54cf45 100644
--- a/proposals/load_ml_model.json
+++ b/proposals/load_ml_model.json
@@ -1,11 +1,8 @@
 {
     "id": "load_ml_model",
     "summary": "Load a ML model",
-    "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit_regr_random_forest()`` and ``save_ml_model()``.",
-    "categories": [
-        "machine learning",
-        "import"
-    ],
+    "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()``.",
+    "categories": ["machine learning", "import"],
     "experimental": true,
     "parameters": [
         {
@@ -43,4 +40,5 @@
             "rel": "about"
         }
     ]
-}
\ No newline at end of file
+}
+
diff --git a/proposals/mlm_regr_random_forest.json b/proposals/mlm_regr_random_forest.json
index 78688f4a..088e2cb0 100644
--- a/proposals/mlm_regr_random_forest.json
+++ b/proposals/mlm_regr_random_forest.json
@@ -55,3 +55,4 @@
         }
     ]
 }
+

From bceb3bf28628a843a171b63dd5b4d6fa084931e2 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Wed, 9 Apr 2025 11:48:58 +0200
Subject: [PATCH 09/52] json formatting

---
 proposals-ml/load_ml_model.json          |  5 ++++-
 proposals-ml/ml_fit.json                 |  4 +++-
 proposals-ml/ml_predict.json             |  4 +++-
 proposals-ml/mlm_regr_random_forest.json | 16 +++++++++++++---
 proposals/load_ml_model.json             |  5 ++++-
 proposals/ml_fit.json                    |  4 +++-
 proposals/ml_predict.json                |  4 +++-
 proposals/mlm_regr_random_forest.json    | 16 +++++++++++++---
 8 files changed, 46 insertions(+), 12 deletions(-)

diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_ml_model.json
index 8a54cf45..28a65660 100644
--- a/proposals-ml/load_ml_model.json
+++ b/proposals-ml/load_ml_model.json
@@ -2,7 +2,10 @@
     "id": "load_ml_model",
     "summary": "Load a ML model",
     "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()``.",
-    "categories": ["machine learning", "import"],
+    "categories": [
+        "machine learning",
+         "import"
+        ],
     "experimental": true,
     "parameters": [
         {
diff --git a/proposals-ml/ml_fit.json b/proposals-ml/ml_fit.json
index cd485e8a..2e01ff04 100644
--- a/proposals-ml/ml_fit.json
+++ b/proposals-ml/ml_fit.json
@@ -2,7 +2,9 @@
     "id": "ml_fit",
     "summary": "Train a machine learning model",
     "description": "Executes the fit of a specified machine learning model based on training data.\n\nThe function is generic and supports different machine learning models.",
-    "categories": ["machine learning"],
+    "categories": [
+        "machine learning"
+    ],
     "experimental": true,
     "parameters": [
         {
diff --git a/proposals-ml/ml_predict.json b/proposals-ml/ml_predict.json
index d090a7a6..3954e95b 100644
--- a/proposals-ml/ml_predict.json
+++ b/proposals-ml/ml_predict.json
@@ -2,7 +2,9 @@
     "id": "ml_predict",
     "summary": "Predict using ML",
     "description": "Applies a machine learning model to a data cube of input features and returns the predicted values.",
-    "categories": ["machine learning"],
+    "categories": [
+        "machine learning"
+    ],
     "experimental": true,
     "parameters": [
         {
diff --git a/proposals-ml/mlm_regr_random_forest.json b/proposals-ml/mlm_regr_random_forest.json
index 088e2cb0..34325c5c 100644
--- a/proposals-ml/mlm_regr_random_forest.json
+++ b/proposals-ml/mlm_regr_random_forest.json
@@ -2,7 +2,9 @@
     "id": "mlm_regr_random_forest",
     "summary": "Initialize a random forest regression model",
     "description": "Initializes a Random Forest regression model. This component prepares the model structure but does not perform training. The model can be trained later using ``ml_fit``.",
-    "categories": ["machine learning"],
+    "categories": [
+        "machine learning"
+    ],
     "experimental": true,
     "parameters": [
         {
@@ -15,7 +17,12 @@
                 },
                 {
                     "type": "string",
-                    "enum": ["all", "log2", "onethird", "sqrt"]
+                    "enum": [
+                        "all", 
+                        "log2", 
+                        "onethird", 
+                        "sqrt"
+                    ]
                 }
             ]
         },
@@ -35,7 +42,10 @@
             "optional": true,
             "default": null,
             "schema": {
-                "type": ["integer", "null"]
+                "type": [
+                    "integer", 
+                    "null"
+                ]
             }
         }
     ],
diff --git a/proposals/load_ml_model.json b/proposals/load_ml_model.json
index 8a54cf45..28a65660 100644
--- a/proposals/load_ml_model.json
+++ b/proposals/load_ml_model.json
@@ -2,7 +2,10 @@
     "id": "load_ml_model",
     "summary": "Load a ML model",
     "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()``.",
-    "categories": ["machine learning", "import"],
+    "categories": [
+        "machine learning",
+         "import"
+        ],
     "experimental": true,
     "parameters": [
         {
diff --git a/proposals/ml_fit.json b/proposals/ml_fit.json
index cd485e8a..2e01ff04 100644
--- a/proposals/ml_fit.json
+++ b/proposals/ml_fit.json
@@ -2,7 +2,9 @@
     "id": "ml_fit",
     "summary": "Train a machine learning model",
     "description": "Executes the fit of a specified machine learning model based on training data.\n\nThe function is generic and supports different machine learning models.",
-    "categories": ["machine learning"],
+    "categories": [
+        "machine learning"
+    ],
     "experimental": true,
     "parameters": [
         {
diff --git a/proposals/ml_predict.json b/proposals/ml_predict.json
index d090a7a6..3954e95b 100644
--- a/proposals/ml_predict.json
+++ b/proposals/ml_predict.json
@@ -2,7 +2,9 @@
     "id": "ml_predict",
     "summary": "Predict using ML",
     "description": "Applies a machine learning model to a data cube of input features and returns the predicted values.",
-    "categories": ["machine learning"],
+    "categories": [
+        "machine learning"
+    ],
     "experimental": true,
     "parameters": [
         {
diff --git a/proposals/mlm_regr_random_forest.json b/proposals/mlm_regr_random_forest.json
index 088e2cb0..34325c5c 100644
--- a/proposals/mlm_regr_random_forest.json
+++ b/proposals/mlm_regr_random_forest.json
@@ -2,7 +2,9 @@
     "id": "mlm_regr_random_forest",
     "summary": "Initialize a random forest regression model",
     "description": "Initializes a Random Forest regression model. This component prepares the model structure but does not perform training. The model can be trained later using ``ml_fit``.",
-    "categories": ["machine learning"],
+    "categories": [
+        "machine learning"
+    ],
     "experimental": true,
     "parameters": [
         {
@@ -15,7 +17,12 @@
                 },
                 {
                     "type": "string",
-                    "enum": ["all", "log2", "onethird", "sqrt"]
+                    "enum": [
+                        "all", 
+                        "log2", 
+                        "onethird", 
+                        "sqrt"
+                    ]
                 }
             ]
         },
@@ -35,7 +42,10 @@
             "optional": true,
             "default": null,
             "schema": {
-                "type": ["integer", "null"]
+                "type": [
+                    "integer", 
+                    "null"
+                ]
             }
         }
     ],

From 0e609b49ced2a3b21ff082fea7a5d654a04ee699 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Wed, 9 Apr 2025 12:00:27 +0200
Subject: [PATCH 10/52] format spaces

---
 proposals-ml/load_ml_model.json          | 4 ++--
 proposals-ml/mlm_regr_random_forest.json | 8 ++++----
 proposals/load_ml_model.json             | 4 ++--
 proposals/mlm_regr_random_forest.json    | 8 ++++----
 4 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_ml_model.json
index 28a65660..4b4e9ed4 100644
--- a/proposals-ml/load_ml_model.json
+++ b/proposals-ml/load_ml_model.json
@@ -4,8 +4,8 @@
     "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()``.",
     "categories": [
         "machine learning",
-         "import"
-        ],
+        "import"
+    ],
     "experimental": true,
     "parameters": [
         {
diff --git a/proposals-ml/mlm_regr_random_forest.json b/proposals-ml/mlm_regr_random_forest.json
index 34325c5c..f643a73a 100644
--- a/proposals-ml/mlm_regr_random_forest.json
+++ b/proposals-ml/mlm_regr_random_forest.json
@@ -18,9 +18,9 @@
                 {
                     "type": "string",
                     "enum": [
-                        "all", 
-                        "log2", 
-                        "onethird", 
+                        "all",
+                        "log2",
+                        "onethird",
                         "sqrt"
                     ]
                 }
@@ -43,7 +43,7 @@
             "default": null,
             "schema": {
                 "type": [
-                    "integer", 
+                    "integer",
                     "null"
                 ]
             }
diff --git a/proposals/load_ml_model.json b/proposals/load_ml_model.json
index 28a65660..4b4e9ed4 100644
--- a/proposals/load_ml_model.json
+++ b/proposals/load_ml_model.json
@@ -4,8 +4,8 @@
     "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()``.",
     "categories": [
         "machine learning",
-         "import"
-        ],
+        "import"
+    ],
     "experimental": true,
     "parameters": [
         {
diff --git a/proposals/mlm_regr_random_forest.json b/proposals/mlm_regr_random_forest.json
index 34325c5c..f643a73a 100644
--- a/proposals/mlm_regr_random_forest.json
+++ b/proposals/mlm_regr_random_forest.json
@@ -18,9 +18,9 @@
                 {
                     "type": "string",
                     "enum": [
-                        "all", 
-                        "log2", 
-                        "onethird", 
+                        "all",
+                        "log2",
+                        "onethird",
                         "sqrt"
                     ]
                 }
@@ -43,7 +43,7 @@
             "default": null,
             "schema": {
                 "type": [
-                    "integer", 
+                    "integer",
                     "null"
                 ]
             }

From 12d28f03f26ff9de786eff8624eff3f46788e148 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Thu, 10 Apr 2025 12:01:30 +0200
Subject: [PATCH 11/52] tempcnn, mlp, xgboost

---
 CHANGELOG.md                        |   3 +
 proposals-ml/mlm_class_mlp.json     | 115 +++++++++++++++++++++++
 proposals-ml/mlm_class_tempcnn.json | 137 ++++++++++++++++++++++++++++
 proposals-ml/mlm_class_xgboost.json |  89 ++++++++++++++++++
 proposals/mlm_class_mlp.json        | 115 +++++++++++++++++++++++
 proposals/mlm_class_tempcnn.json    | 137 ++++++++++++++++++++++++++++
 proposals/mlm_class_xgboost.json    |  89 ++++++++++++++++++
 tests/.words                        |  15 +++
 8 files changed, 700 insertions(+)
 create mode 100644 proposals-ml/mlm_class_mlp.json
 create mode 100644 proposals-ml/mlm_class_tempcnn.json
 create mode 100644 proposals-ml/mlm_class_xgboost.json
 create mode 100644 proposals/mlm_class_mlp.json
 create mode 100644 proposals/mlm_class_tempcnn.json
 create mode 100644 proposals/mlm_class_xgboost.json

diff --git a/CHANGELOG.md b/CHANGELOG.md
index f73ac5c8..07c529e2 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -18,7 +18,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `load_geojson`
     - `load_ml_model`
     - `load_url`
+    - `mlm_class_mlp`
     - `mlm_class_random_forest`
+    - `mlm_class_tempcnn`
+    - `mlm_class_xgboost`
     - `mlm_regr_random_forest`
     - `ml_fit`
     - `ml_predict`
diff --git a/proposals-ml/mlm_class_mlp.json b/proposals-ml/mlm_class_mlp.json
new file mode 100644
index 00000000..d834656b
--- /dev/null
+++ b/proposals-ml/mlm_class_mlp.json
@@ -0,0 +1,115 @@
+{
+    "id": "mlm_class_mlp",
+    "summary": "Initialize a Multi-Layer Perceptron (MLP) classification model",
+    "description": "Initializes a Multi-Layer Perceptron (MLP) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "hidden_layer_sizes",
+            "description": "List of integers specifying the number of neurons in each hidden layer.",
+            "default": [512, 512, 512],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 1
+            }
+        },
+        {
+            "name": "activation",
+            "description": "Activation function for the hidden layers.",
+            "optional": true,
+            "default": "relu",
+            "schema": {
+                "type": "string",
+                "enum": ["relu", "tanh", "logistic"]
+            }
+        },
+        {
+            "name": "dropout_rates",
+            "description": "List of numbers between 0 and 1 specifying the dropout rate for each hidden layer.",
+            "default": [0.4, 0.3, 0.2],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "number",
+                    "minimum": 0,
+                    "maximum": 1
+                },
+                "minItems": 1
+            }
+        },
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 100,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 64,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": ["adam", "adabound", "adabelief", "madagrad", "nadam", "qhadam", "radam", "swats", "yogi"]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://arxiv.org/abs/1611.06455",
+            "title": "Z. Wang, W. Yan, and T. Oates (2017), Time Series Classification from Scratch with Deep Neural Networks: A Strong Baseline",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_tempcnn.json b/proposals-ml/mlm_class_tempcnn.json
new file mode 100644
index 00000000..3584c4a1
--- /dev/null
+++ b/proposals-ml/mlm_class_tempcnn.json
@@ -0,0 +1,137 @@
+{
+    "id": "mlm_class_tempcnn",
+    "summary": "Initialize a Temporal Convolutional Neural Network (TempCNN) classification model",
+    "description": "Initializes a Temporal Convolutional Neural Network (TempCNN) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "cnn_layers",
+            "description": "List of integers specifying the number of filters in each convolutional layer.",
+            "default": [256, 256, 256],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 1
+            }
+        },
+        {
+            "name": "cnn_kernels",
+            "description": "List of integers specifying the kernel size for each convolutional layer.",
+            "default": [7, 7, 7],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 1
+            }
+        },
+        {
+            "name": "cnn_dropout_rates",
+            "description": "List of numbers between 0 and 1 specifying the dropout rate for each convolutional layer.",
+            "default": [0.2, 0.2, 0.2],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "number",
+                    "minimum": 0,
+                    "maximum": 1
+                },
+                "minItems": 1
+            }
+        },
+        {
+            "name": "dense_layer_nodes",
+            "description": "Number of nodes in the dense layer.",
+            "default": 256,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "dense_layer_dropout_rate",
+            "description": "Dropout rate for the dense layer.",
+            "default": 0.5,
+            "schema": {
+                "type": "number",
+                "minimum": 0,
+                "maximum": 1
+            }
+        },
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 100,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 64,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": ["adam", "adabound", "adabelief", "madagrad", "nadam", "qhadam", "radam", "swats", "yogi"]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.3390/rs11050523",
+            "title": "Pelletier et al. (2019): Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
diff --git a/proposals-ml/mlm_class_xgboost.json b/proposals-ml/mlm_class_xgboost.json
new file mode 100644
index 00000000..a7b21f50
--- /dev/null
+++ b/proposals-ml/mlm_class_xgboost.json
@@ -0,0 +1,89 @@
+{
+    "id": "mlm_class_xgboost",
+    "summary": "Initialize an XGBoost classification model",
+    "description": "Initializes an XGBoost classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "learning_rate",
+            "description": "Step size shrinkage used in update to prevent overfitting.",
+            "optional": true,
+            "default": 0.15,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "max_depth",
+            "description": "Maximum depth of a tree.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "min_child_weight",
+            "description": "Minimum sum of instance weight (hessian) needed in a child.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "subsample",
+            "description": "Subsample ratio of the training instance.",
+            "optional": true,
+            "default": 0.8,
+            "schema": {
+                "type": "number",
+                "minimum": 0,
+                "maximum": 1
+            }
+        },
+        {
+            "name": "min_split_loss",
+            "description": "Minimum loss reduction required to make a further partition on a leaf node of the tree.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://dl.acm.org/doi/10.1145/2939672.2939785",
+            "title": "Chen and Guestrin (2016), XGBoost: A Scalable Tree Boosting System",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
\ No newline at end of file
diff --git a/proposals/mlm_class_mlp.json b/proposals/mlm_class_mlp.json
new file mode 100644
index 00000000..d834656b
--- /dev/null
+++ b/proposals/mlm_class_mlp.json
@@ -0,0 +1,115 @@
+{
+    "id": "mlm_class_mlp",
+    "summary": "Initialize a Multi-Layer Perceptron (MLP) classification model",
+    "description": "Initializes a Multi-Layer Perceptron (MLP) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "hidden_layer_sizes",
+            "description": "List of integers specifying the number of neurons in each hidden layer.",
+            "default": [512, 512, 512],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 1
+            }
+        },
+        {
+            "name": "activation",
+            "description": "Activation function for the hidden layers.",
+            "optional": true,
+            "default": "relu",
+            "schema": {
+                "type": "string",
+                "enum": ["relu", "tanh", "logistic"]
+            }
+        },
+        {
+            "name": "dropout_rates",
+            "description": "List of numbers between 0 and 1 specifying the dropout rate for each hidden layer.",
+            "default": [0.4, 0.3, 0.2],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "number",
+                    "minimum": 0,
+                    "maximum": 1
+                },
+                "minItems": 1
+            }
+        },
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 100,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 64,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": ["adam", "adabound", "adabelief", "madagrad", "nadam", "qhadam", "radam", "swats", "yogi"]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://arxiv.org/abs/1611.06455",
+            "title": "Z. Wang, W. Yan, and T. Oates (2017), Time Series Classification from Scratch with Deep Neural Networks: A Strong Baseline",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
\ No newline at end of file
diff --git a/proposals/mlm_class_tempcnn.json b/proposals/mlm_class_tempcnn.json
new file mode 100644
index 00000000..3584c4a1
--- /dev/null
+++ b/proposals/mlm_class_tempcnn.json
@@ -0,0 +1,137 @@
+{
+    "id": "mlm_class_tempcnn",
+    "summary": "Initialize a Temporal Convolutional Neural Network (TempCNN) classification model",
+    "description": "Initializes a Temporal Convolutional Neural Network (TempCNN) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "cnn_layers",
+            "description": "List of integers specifying the number of filters in each convolutional layer.",
+            "default": [256, 256, 256],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 1
+            }
+        },
+        {
+            "name": "cnn_kernels",
+            "description": "List of integers specifying the kernel size for each convolutional layer.",
+            "default": [7, 7, 7],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 1
+            }
+        },
+        {
+            "name": "cnn_dropout_rates",
+            "description": "List of numbers between 0 and 1 specifying the dropout rate for each convolutional layer.",
+            "default": [0.2, 0.2, 0.2],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "number",
+                    "minimum": 0,
+                    "maximum": 1
+                },
+                "minItems": 1
+            }
+        },
+        {
+            "name": "dense_layer_nodes",
+            "description": "Number of nodes in the dense layer.",
+            "default": 256,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "dense_layer_dropout_rate",
+            "description": "Dropout rate for the dense layer.",
+            "default": 0.5,
+            "schema": {
+                "type": "number",
+                "minimum": 0,
+                "maximum": 1
+            }
+        },
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 100,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 64,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": ["adam", "adabound", "adabelief", "madagrad", "nadam", "qhadam", "radam", "swats", "yogi"]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.3390/rs11050523",
+            "title": "Pelletier et al. (2019): Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
diff --git a/proposals/mlm_class_xgboost.json b/proposals/mlm_class_xgboost.json
new file mode 100644
index 00000000..a7b21f50
--- /dev/null
+++ b/proposals/mlm_class_xgboost.json
@@ -0,0 +1,89 @@
+{
+    "id": "mlm_class_xgboost",
+    "summary": "Initialize an XGBoost classification model",
+    "description": "Initializes an XGBoost classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "learning_rate",
+            "description": "Step size shrinkage used in update to prevent overfitting.",
+            "optional": true,
+            "default": 0.15,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "max_depth",
+            "description": "Maximum depth of a tree.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "min_child_weight",
+            "description": "Minimum sum of instance weight (hessian) needed in a child.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "subsample",
+            "description": "Subsample ratio of the training instance.",
+            "optional": true,
+            "default": 0.8,
+            "schema": {
+                "type": "number",
+                "minimum": 0,
+                "maximum": 1
+            }
+        },
+        {
+            "name": "min_split_loss",
+            "description": "Minimum loss reduction required to make a further partition on a leaf node of the tree.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://dl.acm.org/doi/10.1145/2939672.2939785",
+            "title": "Chen and Guestrin (2016), XGBoost: A Scalable Tree Boosting System",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
\ No newline at end of file
diff --git a/tests/.words b/tests/.words
index 2d4e623a..28b1eeed 100644
--- a/tests/.words
+++ b/tests/.words
@@ -49,3 +49,18 @@ date2
 favor
 mlm-model
 analyzes
+XGBoost
+Chen
+Guestrin
+Subsample
+hessian
+overfitting
+Wang
+Yan
+Oates
+adam
+sgd
+minibatches
+Perceptron
+feedforward
+backpropagation

From a3d024b087e9ce5248f431ae806e2f7ea3197ffc Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Thu, 10 Apr 2025 12:15:21 +0200
Subject: [PATCH 12/52] update tempcnn, mlp

---
 proposals-ml/mlm_class_mlp.json     | 32 ++++++++++++++++++++++++-----
 proposals-ml/mlm_class_tempcnn.json | 32 ++++++++++++++++++++++++-----
 tests/.words                        |  1 +
 3 files changed, 55 insertions(+), 10 deletions(-)

diff --git a/proposals-ml/mlm_class_mlp.json b/proposals-ml/mlm_class_mlp.json
index d834656b..45722c74 100644
--- a/proposals-ml/mlm_class_mlp.json
+++ b/proposals-ml/mlm_class_mlp.json
@@ -1,6 +1,6 @@
 {
     "id": "mlm_class_mlp",
-    "summary": "Initialize a Multi-Layer Perceptron (MLP) classification model",
+    "summary": "Initialize an MLP classification model",
     "description": "Initializes a Multi-Layer Perceptron (MLP) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
     "categories": [
         "machine learning"
@@ -10,7 +10,11 @@
         {
             "name": "hidden_layer_sizes",
             "description": "List of integers specifying the number of neurons in each hidden layer.",
-            "default": [512, 512, 512],
+            "default": [
+                512,
+                512,
+                512
+            ],
             "schema": {
                 "type": "array",
                 "items": {
@@ -27,13 +31,21 @@
             "default": "relu",
             "schema": {
                 "type": "string",
-                "enum": ["relu", "tanh", "logistic"]
+                "enum": [
+                    "relu",
+                    "tanh",
+                    "logistic"
+                ]
             }
         },
         {
             "name": "dropout_rates",
             "description": "List of numbers between 0 and 1 specifying the dropout rate for each hidden layer.",
-            "default": [0.4, 0.3, 0.2],
+            "default": [
+                0.4,
+                0.3,
+                0.2
+            ],
             "schema": {
                 "type": "array",
                 "items": {
@@ -71,7 +83,17 @@
             "default": "adam",
             "schema": {
                 "type": "string",
-                "enum": ["adam", "adabound", "adabelief", "madagrad", "nadam", "qhadam", "radam", "swats", "yogi"]
+                "enum": [
+                    "adam",
+                    "adabound",
+                    "adabelief",
+                    "madagrad",
+                    "nadam",
+                    "qhadam",
+                    "radam",
+                    "swats",
+                    "yogi"
+                ]
             }
         },
         {
diff --git a/proposals-ml/mlm_class_tempcnn.json b/proposals-ml/mlm_class_tempcnn.json
index 3584c4a1..cc9da6f6 100644
--- a/proposals-ml/mlm_class_tempcnn.json
+++ b/proposals-ml/mlm_class_tempcnn.json
@@ -1,6 +1,6 @@
 {
     "id": "mlm_class_tempcnn",
-    "summary": "Initialize a Temporal Convolutional Neural Network (TempCNN) classification model",
+    "summary": "Initialize a TempCNN classification model",
     "description": "Initializes a Temporal Convolutional Neural Network (TempCNN) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
     "categories": [
         "machine learning"
@@ -10,7 +10,11 @@
         {
             "name": "cnn_layers",
             "description": "List of integers specifying the number of filters in each convolutional layer.",
-            "default": [256, 256, 256],
+            "default": [
+                256,
+                256,
+                256
+            ],
             "schema": {
                 "type": "array",
                 "items": {
@@ -23,7 +27,11 @@
         {
             "name": "cnn_kernels",
             "description": "List of integers specifying the kernel size for each convolutional layer.",
-            "default": [7, 7, 7],
+            "default": [
+                7,
+                7,
+                7
+            ],
             "schema": {
                 "type": "array",
                 "items": {
@@ -36,7 +44,11 @@
         {
             "name": "cnn_dropout_rates",
             "description": "List of numbers between 0 and 1 specifying the dropout rate for each convolutional layer.",
-            "default": [0.2, 0.2, 0.2],
+            "default": [
+                0.2,
+                0.2,
+                0.2
+            ],
             "schema": {
                 "type": "array",
                 "items": {
@@ -93,7 +105,17 @@
             "default": "adam",
             "schema": {
                 "type": "string",
-                "enum": ["adam", "adabound", "adabelief", "madagrad", "nadam", "qhadam", "radam", "swats", "yogi"]
+                "enum": [
+                    "adam",
+                    "adabound",
+                    "adabelief",
+                    "madagrad",
+                    "nadam",
+                    "qhadam",
+                    "radam",
+                    "swats",
+                    "yogi"
+                ]
             }
         },
         {
diff --git a/tests/.words b/tests/.words
index 28b1eeed..9fbe8d9d 100644
--- a/tests/.words
+++ b/tests/.words
@@ -64,3 +64,4 @@ minibatches
 Perceptron
 feedforward
 backpropagation
+TempCNN

From d84a074458e580c16145bcd20bf92ad95d03ee87 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Thu, 10 Apr 2025 12:19:57 +0200
Subject: [PATCH 13/52] update tempcnn, mlp

---
 proposals/mlm_class_mlp.json     | 32 +++++++++++++++++++++++++++-----
 proposals/mlm_class_tempcnn.json | 32 +++++++++++++++++++++++++++-----
 tests/.words                     |  1 +
 3 files changed, 55 insertions(+), 10 deletions(-)

diff --git a/proposals/mlm_class_mlp.json b/proposals/mlm_class_mlp.json
index d834656b..45722c74 100644
--- a/proposals/mlm_class_mlp.json
+++ b/proposals/mlm_class_mlp.json
@@ -1,6 +1,6 @@
 {
     "id": "mlm_class_mlp",
-    "summary": "Initialize a Multi-Layer Perceptron (MLP) classification model",
+    "summary": "Initialize an MLP classification model",
     "description": "Initializes a Multi-Layer Perceptron (MLP) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
     "categories": [
         "machine learning"
@@ -10,7 +10,11 @@
         {
             "name": "hidden_layer_sizes",
             "description": "List of integers specifying the number of neurons in each hidden layer.",
-            "default": [512, 512, 512],
+            "default": [
+                512,
+                512,
+                512
+            ],
             "schema": {
                 "type": "array",
                 "items": {
@@ -27,13 +31,21 @@
             "default": "relu",
             "schema": {
                 "type": "string",
-                "enum": ["relu", "tanh", "logistic"]
+                "enum": [
+                    "relu",
+                    "tanh",
+                    "logistic"
+                ]
             }
         },
         {
             "name": "dropout_rates",
             "description": "List of numbers between 0 and 1 specifying the dropout rate for each hidden layer.",
-            "default": [0.4, 0.3, 0.2],
+            "default": [
+                0.4,
+                0.3,
+                0.2
+            ],
             "schema": {
                 "type": "array",
                 "items": {
@@ -71,7 +83,17 @@
             "default": "adam",
             "schema": {
                 "type": "string",
-                "enum": ["adam", "adabound", "adabelief", "madagrad", "nadam", "qhadam", "radam", "swats", "yogi"]
+                "enum": [
+                    "adam",
+                    "adabound",
+                    "adabelief",
+                    "madagrad",
+                    "nadam",
+                    "qhadam",
+                    "radam",
+                    "swats",
+                    "yogi"
+                ]
             }
         },
         {
diff --git a/proposals/mlm_class_tempcnn.json b/proposals/mlm_class_tempcnn.json
index 3584c4a1..cc9da6f6 100644
--- a/proposals/mlm_class_tempcnn.json
+++ b/proposals/mlm_class_tempcnn.json
@@ -1,6 +1,6 @@
 {
     "id": "mlm_class_tempcnn",
-    "summary": "Initialize a Temporal Convolutional Neural Network (TempCNN) classification model",
+    "summary": "Initialize a TempCNN classification model",
     "description": "Initializes a Temporal Convolutional Neural Network (TempCNN) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
     "categories": [
         "machine learning"
@@ -10,7 +10,11 @@
         {
             "name": "cnn_layers",
             "description": "List of integers specifying the number of filters in each convolutional layer.",
-            "default": [256, 256, 256],
+            "default": [
+                256,
+                256,
+                256
+            ],
             "schema": {
                 "type": "array",
                 "items": {
@@ -23,7 +27,11 @@
         {
             "name": "cnn_kernels",
             "description": "List of integers specifying the kernel size for each convolutional layer.",
-            "default": [7, 7, 7],
+            "default": [
+                7,
+                7,
+                7
+            ],
             "schema": {
                 "type": "array",
                 "items": {
@@ -36,7 +44,11 @@
         {
             "name": "cnn_dropout_rates",
             "description": "List of numbers between 0 and 1 specifying the dropout rate for each convolutional layer.",
-            "default": [0.2, 0.2, 0.2],
+            "default": [
+                0.2,
+                0.2,
+                0.2
+            ],
             "schema": {
                 "type": "array",
                 "items": {
@@ -93,7 +105,17 @@
             "default": "adam",
             "schema": {
                 "type": "string",
-                "enum": ["adam", "adabound", "adabelief", "madagrad", "nadam", "qhadam", "radam", "swats", "yogi"]
+                "enum": [
+                    "adam",
+                    "adabound",
+                    "adabelief",
+                    "madagrad",
+                    "nadam",
+                    "qhadam",
+                    "radam",
+                    "swats",
+                    "yogi"
+                ]
             }
         },
         {
diff --git a/tests/.words b/tests/.words
index 9fbe8d9d..9bec28b3 100644
--- a/tests/.words
+++ b/tests/.words
@@ -65,3 +65,4 @@ Perceptron
 feedforward
 backpropagation
 TempCNN
+Pelletier

From e423ea46d198ac1794819fe5205e98a169657f80 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Fri, 11 Apr 2025 21:12:16 +0200
Subject: [PATCH 14/52] tae, svm, lighttae, predict probabilities

---
 CHANGELOG.md                               |   4 +
 proposals-ml/ml_predict_probabilities.json |  49 ++++++++
 proposals-ml/mlm_class_lighttae.json       | 128 +++++++++++++++++++++
 proposals-ml/mlm_class_svm.json            | 112 ++++++++++++++++++
 proposals-ml/mlm_class_tae.json            | 128 +++++++++++++++++++++
 proposals/ml_predict_probabilities.json    |  49 ++++++++
 proposals/mlm_class_lighttae.json          | 128 +++++++++++++++++++++
 proposals/mlm_class_svm.json               | 112 ++++++++++++++++++
 proposals/mlm_class_tae.json               | 128 +++++++++++++++++++++
 tests/.words                               |   8 ++
 10 files changed, 846 insertions(+)
 create mode 100644 proposals-ml/ml_predict_probabilities.json
 create mode 100644 proposals-ml/mlm_class_lighttae.json
 create mode 100644 proposals-ml/mlm_class_svm.json
 create mode 100644 proposals-ml/mlm_class_tae.json
 create mode 100644 proposals/ml_predict_probabilities.json
 create mode 100644 proposals/mlm_class_lighttae.json
 create mode 100644 proposals/mlm_class_svm.json
 create mode 100644 proposals/mlm_class_tae.json

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 07c529e2..fb2e6a3f 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -18,13 +18,17 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `load_geojson`
     - `load_ml_model`
     - `load_url`
+    - `mlm_class_lighttae`
     - `mlm_class_mlp`
     - `mlm_class_random_forest`
+    - `mlm_class_svm`
+    - `mlm_class_tae`
     - `mlm_class_tempcnn`
     - `mlm_class_xgboost`
     - `mlm_regr_random_forest`
     - `ml_fit`
     - `ml_predict`
+    - `ml_predict_probabilities`
     - `save_ml_model`
     - `unflatten_dimension`
     - `vector_buffer`
diff --git a/proposals-ml/ml_predict_probabilities.json b/proposals-ml/ml_predict_probabilities.json
new file mode 100644
index 00000000..68cc78c7
--- /dev/null
+++ b/proposals-ml/ml_predict_probabilities.json
@@ -0,0 +1,49 @@
+{
+    "id": "ml_predict_probabilities",
+    "summary": "Predict class probabilities using ML",
+    "description": "Applies a machine learning model to a data cube of input features and returns the predicted class probabilities.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The data cube containing the input features.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "model",
+            "description": "A ML model that was trained and supports probability predictions.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "dimensions",
+            "description": "Zero or more dimensions that will be reduced by the model. Fails with a `DimensionNotAvailable` exception if one of the specified dimensions does not exist.",
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "string"
+                }
+            }
+        }
+    ],
+    "returns": {
+        "description": "A data cube with the predicted class probabilities. It removes the specified dimensions and adds a new dimension for the class probabilities. The dimension has the name `classes` and is of type `other`. Each label in the dimension represents a class, and the values are the probabilities for each class.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube",
+            "dimensions": [
+                {
+                    "type": "other"
+                }
+            ]
+        }
+    }
+}
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_lighttae.json b/proposals-ml/mlm_class_lighttae.json
new file mode 100644
index 00000000..e03e18c0
--- /dev/null
+++ b/proposals-ml/mlm_class_lighttae.json
@@ -0,0 +1,128 @@
+{
+    "id": "mlm_class_lighttae",
+    "summary": "Initialize a LightTAE classification model",
+    "description": "Initializes a Lightweight Temporal Self-Attention (LightTAE) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 150,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 128,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "adam",
+                    "adabound",
+                    "adabelief",
+                    "madagrad",
+                    "nadam",
+                    "qhadam",
+                    "radam",
+                    "swats",
+                    "yogi"
+                ]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.0005,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "epsilon",
+            "description": "A small constant for numerical stability in the optimizer.",
+            "optional": true,
+            "default": 0.00000001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "weight_decay",
+            "description": "Weight decay (L2 penalty) for the optimizer.",
+            "optional": true,
+            "default": 0.0007,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "lr_decay_epochs",
+            "description": "Number of epochs between learning rate decay steps.",
+            "optional": true,
+            "default": 50,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "lr_decay_rate",
+            "description": "Rate at which the learning rate decays.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.1007/978-3-030-65742-0_12",
+            "title": "V. S. F. Garnot and L. Landrieu (2020), Lightweight Temporal Self-attention for Classifying Satellite Images Time Series",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_svm.json b/proposals-ml/mlm_class_svm.json
new file mode 100644
index 00000000..bf0a8ca7
--- /dev/null
+++ b/proposals-ml/mlm_class_svm.json
@@ -0,0 +1,112 @@
+{
+    "id": "mlm_class_svm",
+    "summary": "Initialize an SVM classification model",
+    "description": "Initializes a Support Vector Machine (SVM) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "kernel",
+            "description": "Specifies the kernel type to be used in the algorithm.",
+            "optional": true,
+            "default": "rbf",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "linear",
+                    "poly",
+                    "rbf",
+                    "sigmoid"
+                ]
+            }
+        },
+        {
+            "name": "C",
+            "description": "Regularization parameter. The strength of the regularization is inversely proportional to C. Must be strictly positive.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "gamma",
+            "description": "Kernel coefficient for 'rbf', 'poly', and 'sigmoid'. Higher values lead to tighter fits.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "degree",
+            "description": "Degree of the polynomial kernel function (only relevant for 'poly' kernel).",
+            "optional": true,
+            "default": 3,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "coef0",
+            "description": "Independent term in the kernel function (only relevant for 'poly' and 'sigmoid' kernels).",
+            "optional": true,
+            "default": 0,
+            "schema": {
+                "type": "number"
+            }
+        },
+        {
+            "name": "tolerance",
+            "description": "Tolerance of termination criterion.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "cachesize",
+            "description": "Size of the kernel cache in MB.",
+            "optional": true,
+            "default": 1000,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://link.springer.com/article/10.1007/BF00994018",
+            "title": "C. Cortes and V. Vapnik (1995), Support-vector networks",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_tae.json b/proposals-ml/mlm_class_tae.json
new file mode 100644
index 00000000..ac310e7a
--- /dev/null
+++ b/proposals-ml/mlm_class_tae.json
@@ -0,0 +1,128 @@
+{
+    "id": "mlm_class_tae",
+    "summary": "Initialize a TAE classification model",
+    "description": "Initializes a Temporal Self-Attention (TAE) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 150,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 64,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "adam",
+                    "adabound",
+                    "adabelief",
+                    "madagrad",
+                    "nadam",
+                    "qhadam",
+                    "radam",
+                    "swats",
+                    "yogi"
+                ]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "epsilon",
+            "description": "A small constant for numerical stability in the optimizer.",
+            "optional": true,
+            "default": 0.00000001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "weight_decay",
+            "description": "Weight decay (L2 penalty) for the optimizer.",
+            "optional": true,
+            "default": 0.000001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "lr_decay_epochs",
+            "description": "Number of epochs between learning rate decay steps.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "lr_decay_rate",
+            "description": "Rate at which the learning rate decays.",
+            "optional": true,
+            "default": 0.95,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.1109/CVPR42600.2020.01234",
+            "title": "V. Garnot, L. Landrieu, S. Giordano, and N. Chehata (2020), Satellite Image Time Series Classification With Pixel-Set Encoders and Temporal Self-Attention",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals/ml_predict_probabilities.json b/proposals/ml_predict_probabilities.json
new file mode 100644
index 00000000..68cc78c7
--- /dev/null
+++ b/proposals/ml_predict_probabilities.json
@@ -0,0 +1,49 @@
+{
+    "id": "ml_predict_probabilities",
+    "summary": "Predict class probabilities using ML",
+    "description": "Applies a machine learning model to a data cube of input features and returns the predicted class probabilities.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The data cube containing the input features.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "model",
+            "description": "A ML model that was trained and supports probability predictions.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "dimensions",
+            "description": "Zero or more dimensions that will be reduced by the model. Fails with a `DimensionNotAvailable` exception if one of the specified dimensions does not exist.",
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "string"
+                }
+            }
+        }
+    ],
+    "returns": {
+        "description": "A data cube with the predicted class probabilities. It removes the specified dimensions and adds a new dimension for the class probabilities. The dimension has the name `classes` and is of type `other`. Each label in the dimension represents a class, and the values are the probabilities for each class.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube",
+            "dimensions": [
+                {
+                    "type": "other"
+                }
+            ]
+        }
+    }
+}
\ No newline at end of file
diff --git a/proposals/mlm_class_lighttae.json b/proposals/mlm_class_lighttae.json
new file mode 100644
index 00000000..e03e18c0
--- /dev/null
+++ b/proposals/mlm_class_lighttae.json
@@ -0,0 +1,128 @@
+{
+    "id": "mlm_class_lighttae",
+    "summary": "Initialize a LightTAE classification model",
+    "description": "Initializes a Lightweight Temporal Self-Attention (LightTAE) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 150,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 128,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "adam",
+                    "adabound",
+                    "adabelief",
+                    "madagrad",
+                    "nadam",
+                    "qhadam",
+                    "radam",
+                    "swats",
+                    "yogi"
+                ]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.0005,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "epsilon",
+            "description": "A small constant for numerical stability in the optimizer.",
+            "optional": true,
+            "default": 0.00000001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "weight_decay",
+            "description": "Weight decay (L2 penalty) for the optimizer.",
+            "optional": true,
+            "default": 0.0007,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "lr_decay_epochs",
+            "description": "Number of epochs between learning rate decay steps.",
+            "optional": true,
+            "default": 50,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "lr_decay_rate",
+            "description": "Rate at which the learning rate decays.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.1007/978-3-030-65742-0_12",
+            "title": "V. S. F. Garnot and L. Landrieu (2020), Lightweight Temporal Self-attention for Classifying Satellite Images Time Series",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals/mlm_class_svm.json b/proposals/mlm_class_svm.json
new file mode 100644
index 00000000..bf0a8ca7
--- /dev/null
+++ b/proposals/mlm_class_svm.json
@@ -0,0 +1,112 @@
+{
+    "id": "mlm_class_svm",
+    "summary": "Initialize an SVM classification model",
+    "description": "Initializes a Support Vector Machine (SVM) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "kernel",
+            "description": "Specifies the kernel type to be used in the algorithm.",
+            "optional": true,
+            "default": "rbf",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "linear",
+                    "poly",
+                    "rbf",
+                    "sigmoid"
+                ]
+            }
+        },
+        {
+            "name": "C",
+            "description": "Regularization parameter. The strength of the regularization is inversely proportional to C. Must be strictly positive.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "gamma",
+            "description": "Kernel coefficient for 'rbf', 'poly', and 'sigmoid'. Higher values lead to tighter fits.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "degree",
+            "description": "Degree of the polynomial kernel function (only relevant for 'poly' kernel).",
+            "optional": true,
+            "default": 3,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "coef0",
+            "description": "Independent term in the kernel function (only relevant for 'poly' and 'sigmoid' kernels).",
+            "optional": true,
+            "default": 0,
+            "schema": {
+                "type": "number"
+            }
+        },
+        {
+            "name": "tolerance",
+            "description": "Tolerance of termination criterion.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "cachesize",
+            "description": "Size of the kernel cache in MB.",
+            "optional": true,
+            "default": 1000,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://link.springer.com/article/10.1007/BF00994018",
+            "title": "C. Cortes and V. Vapnik (1995), Support-vector networks",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
\ No newline at end of file
diff --git a/proposals/mlm_class_tae.json b/proposals/mlm_class_tae.json
new file mode 100644
index 00000000..ac310e7a
--- /dev/null
+++ b/proposals/mlm_class_tae.json
@@ -0,0 +1,128 @@
+{
+    "id": "mlm_class_tae",
+    "summary": "Initialize a TAE classification model",
+    "description": "Initializes a Temporal Self-Attention (TAE) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 150,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 64,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "adam",
+                    "adabound",
+                    "adabelief",
+                    "madagrad",
+                    "nadam",
+                    "qhadam",
+                    "radam",
+                    "swats",
+                    "yogi"
+                ]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "epsilon",
+            "description": "A small constant for numerical stability in the optimizer.",
+            "optional": true,
+            "default": 0.00000001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "weight_decay",
+            "description": "Weight decay (L2 penalty) for the optimizer.",
+            "optional": true,
+            "default": 0.000001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "lr_decay_epochs",
+            "description": "Number of epochs between learning rate decay steps.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "lr_decay_rate",
+            "description": "Rate at which the learning rate decays.",
+            "optional": true,
+            "default": 0.95,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.1109/CVPR42600.2020.01234",
+            "title": "V. Garnot, L. Landrieu, S. Giordano, and N. Chehata (2020), Satellite Image Time Series Classification With Pixel-Set Encoders and Temporal Self-Attention",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/tests/.words b/tests/.words
index 9bec28b3..41c2de8e 100644
--- a/tests/.words
+++ b/tests/.words
@@ -66,3 +66,11 @@ feedforward
 backpropagation
 TempCNN
 Pelletier
+Cortes
+Vapnik
+rbf
+SVM
+Garnot
+Landrieu
+Giordano
+Chehata

From 3758fd9359e99f592ff6a8749cb306c8bdf51a69 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Fri, 11 Apr 2025 21:21:43 +0200
Subject: [PATCH 15/52] LightTAE format

---
 proposals-ml/mlm_class_lighttae.json | 2 +-
 proposals-ml/mlm_class_tae.json      | 2 +-
 proposals/mlm_class_lighttae.json    | 2 +-
 proposals/mlm_class_tae.json         | 2 +-
 tests/.words                         | 2 ++
 5 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/proposals-ml/mlm_class_lighttae.json b/proposals-ml/mlm_class_lighttae.json
index e03e18c0..b9f88533 100644
--- a/proposals-ml/mlm_class_lighttae.json
+++ b/proposals-ml/mlm_class_lighttae.json
@@ -61,7 +61,7 @@
             "name": "epsilon",
             "description": "A small constant for numerical stability in the optimizer.",
             "optional": true,
-            "default": 0.00000001,
+            "default": 1e-8,
             "schema": {
                 "type": "number",
                 "minimum": 0
diff --git a/proposals-ml/mlm_class_tae.json b/proposals-ml/mlm_class_tae.json
index ac310e7a..6f81e30e 100644
--- a/proposals-ml/mlm_class_tae.json
+++ b/proposals-ml/mlm_class_tae.json
@@ -61,7 +61,7 @@
             "name": "epsilon",
             "description": "A small constant for numerical stability in the optimizer.",
             "optional": true,
-            "default": 0.00000001,
+            "default": 1e-8,
             "schema": {
                 "type": "number",
                 "minimum": 0
diff --git a/proposals/mlm_class_lighttae.json b/proposals/mlm_class_lighttae.json
index e03e18c0..b9f88533 100644
--- a/proposals/mlm_class_lighttae.json
+++ b/proposals/mlm_class_lighttae.json
@@ -61,7 +61,7 @@
             "name": "epsilon",
             "description": "A small constant for numerical stability in the optimizer.",
             "optional": true,
-            "default": 0.00000001,
+            "default": 1e-8,
             "schema": {
                 "type": "number",
                 "minimum": 0
diff --git a/proposals/mlm_class_tae.json b/proposals/mlm_class_tae.json
index ac310e7a..6f81e30e 100644
--- a/proposals/mlm_class_tae.json
+++ b/proposals/mlm_class_tae.json
@@ -61,7 +61,7 @@
             "name": "epsilon",
             "description": "A small constant for numerical stability in the optimizer.",
             "optional": true,
-            "default": 0.00000001,
+            "default": 1e-8,
             "schema": {
                 "type": "number",
                 "minimum": 0
diff --git a/tests/.words b/tests/.words
index 41c2de8e..8e601773 100644
--- a/tests/.words
+++ b/tests/.words
@@ -74,3 +74,5 @@ Garnot
 Landrieu
 Giordano
 Chehata
+LightTAE
+TAE

From 48fb88bc463c6adfe7f4779cfa357da29c8ce91a Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Tue, 15 Apr 2025 16:20:04 +0200
Subject: [PATCH 16/52] export, import, cube, ml_model

---
 CHANGELOG.md                      |  4 ++++
 proposals-ml/export_cube.json     | 40 +++++++++++++++++++++++++++++++
 proposals-ml/export_ml_model.json | 40 +++++++++++++++++++++++++++++++
 proposals-ml/import_cube.json     | 33 +++++++++++++++++++++++++
 proposals-ml/import_ml_model.json | 33 +++++++++++++++++++++++++
 proposals/export_cube.json        | 40 +++++++++++++++++++++++++++++++
 proposals/export_ml_model.json    | 40 +++++++++++++++++++++++++++++++
 proposals/import_cube.json        | 33 +++++++++++++++++++++++++
 proposals/import_ml_model.json    | 33 +++++++++++++++++++++++++
 9 files changed, 296 insertions(+)
 create mode 100644 proposals-ml/export_cube.json
 create mode 100644 proposals-ml/export_ml_model.json
 create mode 100644 proposals-ml/import_cube.json
 create mode 100644 proposals-ml/import_ml_model.json
 create mode 100644 proposals/export_cube.json
 create mode 100644 proposals/export_ml_model.json
 create mode 100644 proposals/import_cube.json
 create mode 100644 proposals/import_ml_model.json

diff --git a/CHANGELOG.md b/CHANGELOG.md
index fb2e6a3f..bffcc5fc 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -13,8 +13,12 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - New processes in proposal state:
     - `date_between`
     - `date_difference`
+    - `export_cube`
+    - `export_ml_model`
     - `filter_vector`
     - `flatten_dimensions`
+    - `import_cube`
+    - `import_ml_model`
     - `load_geojson`
     - `load_ml_model`
     - `load_url`
diff --git a/proposals-ml/export_cube.json b/proposals-ml/export_cube.json
new file mode 100644
index 00000000..a1292b0e
--- /dev/null
+++ b/proposals-ml/export_cube.json
@@ -0,0 +1,40 @@
+{
+    "id": "export_cube",
+    "summary": "Export a data cube to a workspace",
+    "description": "Exports a processed data cube to a specified workspace location. This allows for storing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) for later use across various machine learning models.\n\nThe workspace is recommended to be on the server backend,this ensures better performance.",
+    "categories": [
+        "Export",
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The data cube to be exported. This should be a processed data cube that you want to store for later use.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "name",
+            "description": "The name to be given to the exported data cube. This will be used to identify the data when importing it later.",
+            "schema": {
+                "type": "string"
+            }
+        },
+        {
+            "name": "folder",
+            "description": "The folder path within the server backend workspace where the data cube should be stored. This helps organize different data cubes within the workspace.",
+            "schema": {
+                "type": "string"
+            }
+        }
+    ],
+    "returns": {
+        "description": "A boolean indicating whether the export was successful.",
+        "schema": {
+            "type": "boolean"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/proposals-ml/export_ml_model.json b/proposals-ml/export_ml_model.json
new file mode 100644
index 00000000..0c6cb781
--- /dev/null
+++ b/proposals-ml/export_ml_model.json
@@ -0,0 +1,40 @@
+{
+    "id": "export_ml_model",
+    "summary": "Export a machine learning model to a workspace",
+    "description": "Exports a machine learning model to a specified workspace location. This is designed for intermediate model storage and can use the backend language's native serialization format.\n\nUnlike save_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for temporary storage and transfer of models between processes.",
+    "categories": [
+        "Export",
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "model",
+            "description": "The machine learning model to be exported. This can be in any format supported by the backend implementation.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "name",
+            "description": "The name to be given to the exported model. This will be used to identify the model when importing it later.",
+            "schema": {
+                "type": "string"
+            }
+        },
+        {
+            "name": "folder",
+            "description": "The folder path within the server backend workspace where the model should be stored.",
+            "schema": {
+                "type": "string"
+            }
+        }
+    ],
+    "returns": {
+        "description": "A boolean indicating whether the export was successful.",
+        "schema": {
+            "type": "boolean"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/proposals-ml/import_cube.json b/proposals-ml/import_cube.json
new file mode 100644
index 00000000..b4d9987a
--- /dev/null
+++ b/proposals-ml/import_cube.json
@@ -0,0 +1,33 @@
+{
+    "id": "import_cube",
+    "summary": "Import a data cube from a workspace",
+    "description": "Imports a previously exported data cube from a specified workspace location. This allows for reusing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) across different machine learning workflows.",
+    "categories": [
+        "Import",
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "name",
+            "description": "The name of the data cube to import. This should match the name used when the data cube was exported.",
+            "schema": {
+                "type": "string"
+            }
+        },
+        {
+            "name": "folder",
+            "description": "The folder path within the server backend workspace where the data cube is stored. This should match the folder path used when the data cube was exported.",
+            "schema": {
+                "type": "string"
+            }
+        }
+    ],
+    "returns": {
+        "description": "The imported data cube that can be used in subsequent processing steps.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/proposals-ml/import_ml_model.json b/proposals-ml/import_ml_model.json
new file mode 100644
index 00000000..ede1888b
--- /dev/null
+++ b/proposals-ml/import_ml_model.json
@@ -0,0 +1,33 @@
+{
+    "id": "import_ml_model",
+    "summary": "Import a machine learning model from a workspace",
+    "description": "Imports a previously exported machine learning model from a specified workspace location. This is designed for intermediate model retrieval and can handle models in the backend language's native serialization format.\n\nUnlike load_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for retrieving temporarily stored models between processes.",
+    "categories": [
+        "Import",
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "name",
+            "description": "The name of the model to import. This should match the name used when the model was exported.",
+            "schema": {
+                "type": "string"
+            }
+        },
+        {
+            "name": "folder",
+            "description": "The folder path within the server backend workspace where the model is stored.",
+            "schema": {
+                "type": "string"
+            }
+        }
+    ],
+    "returns": {
+        "description": "The imported machine learning model in its original format.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/proposals/export_cube.json b/proposals/export_cube.json
new file mode 100644
index 00000000..a1292b0e
--- /dev/null
+++ b/proposals/export_cube.json
@@ -0,0 +1,40 @@
+{
+    "id": "export_cube",
+    "summary": "Export a data cube to a workspace",
+    "description": "Exports a processed data cube to a specified workspace location. This allows for storing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) for later use across various machine learning models.\n\nThe workspace is recommended to be on the server backend,this ensures better performance.",
+    "categories": [
+        "Export",
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The data cube to be exported. This should be a processed data cube that you want to store for later use.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "name",
+            "description": "The name to be given to the exported data cube. This will be used to identify the data when importing it later.",
+            "schema": {
+                "type": "string"
+            }
+        },
+        {
+            "name": "folder",
+            "description": "The folder path within the server backend workspace where the data cube should be stored. This helps organize different data cubes within the workspace.",
+            "schema": {
+                "type": "string"
+            }
+        }
+    ],
+    "returns": {
+        "description": "A boolean indicating whether the export was successful.",
+        "schema": {
+            "type": "boolean"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/proposals/export_ml_model.json b/proposals/export_ml_model.json
new file mode 100644
index 00000000..0c6cb781
--- /dev/null
+++ b/proposals/export_ml_model.json
@@ -0,0 +1,40 @@
+{
+    "id": "export_ml_model",
+    "summary": "Export a machine learning model to a workspace",
+    "description": "Exports a machine learning model to a specified workspace location. This is designed for intermediate model storage and can use the backend language's native serialization format.\n\nUnlike save_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for temporary storage and transfer of models between processes.",
+    "categories": [
+        "Export",
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "model",
+            "description": "The machine learning model to be exported. This can be in any format supported by the backend implementation.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "name",
+            "description": "The name to be given to the exported model. This will be used to identify the model when importing it later.",
+            "schema": {
+                "type": "string"
+            }
+        },
+        {
+            "name": "folder",
+            "description": "The folder path within the server backend workspace where the model should be stored.",
+            "schema": {
+                "type": "string"
+            }
+        }
+    ],
+    "returns": {
+        "description": "A boolean indicating whether the export was successful.",
+        "schema": {
+            "type": "boolean"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/proposals/import_cube.json b/proposals/import_cube.json
new file mode 100644
index 00000000..b4d9987a
--- /dev/null
+++ b/proposals/import_cube.json
@@ -0,0 +1,33 @@
+{
+    "id": "import_cube",
+    "summary": "Import a data cube from a workspace",
+    "description": "Imports a previously exported data cube from a specified workspace location. This allows for reusing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) across different machine learning workflows.",
+    "categories": [
+        "Import",
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "name",
+            "description": "The name of the data cube to import. This should match the name used when the data cube was exported.",
+            "schema": {
+                "type": "string"
+            }
+        },
+        {
+            "name": "folder",
+            "description": "The folder path within the server backend workspace where the data cube is stored. This should match the folder path used when the data cube was exported.",
+            "schema": {
+                "type": "string"
+            }
+        }
+    ],
+    "returns": {
+        "description": "The imported data cube that can be used in subsequent processing steps.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/proposals/import_ml_model.json b/proposals/import_ml_model.json
new file mode 100644
index 00000000..ede1888b
--- /dev/null
+++ b/proposals/import_ml_model.json
@@ -0,0 +1,33 @@
+{
+    "id": "import_ml_model",
+    "summary": "Import a machine learning model from a workspace",
+    "description": "Imports a previously exported machine learning model from a specified workspace location. This is designed for intermediate model retrieval and can handle models in the backend language's native serialization format.\n\nUnlike load_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for retrieving temporarily stored models between processes.",
+    "categories": [
+        "Import",
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "name",
+            "description": "The name of the model to import. This should match the name used when the model was exported.",
+            "schema": {
+                "type": "string"
+            }
+        },
+        {
+            "name": "folder",
+            "description": "The folder path within the server backend workspace where the model is stored.",
+            "schema": {
+                "type": "string"
+            }
+        }
+    ],
+    "returns": {
+        "description": "The imported machine learning model in its original format.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    }
+} 
\ No newline at end of file

From ef51a9c75e06ddecfecac2a978431d67ca19b8c7 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Tue, 15 Apr 2025 16:31:10 +0200
Subject: [PATCH 17/52] resolve naming - backend

---
 proposals-ml/export_cube.json     | 4 ++--
 proposals-ml/export_ml_model.json | 6 +++---
 proposals-ml/import_cube.json     | 2 +-
 proposals-ml/import_ml_model.json | 4 ++--
 proposals/export_cube.json        | 4 ++--
 proposals/export_ml_model.json    | 6 +++---
 proposals/import_cube.json        | 2 +-
 proposals/import_ml_model.json    | 4 ++--
 8 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/proposals-ml/export_cube.json b/proposals-ml/export_cube.json
index a1292b0e..8810c653 100644
--- a/proposals-ml/export_cube.json
+++ b/proposals-ml/export_cube.json
@@ -1,7 +1,7 @@
 {
     "id": "export_cube",
     "summary": "Export a data cube to a workspace",
-    "description": "Exports a processed data cube to a specified workspace location. This allows for storing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) for later use across various machine learning models.\n\nThe workspace is recommended to be on the server backend,this ensures better performance.",
+    "description": "Exports a processed data cube to a specified workspace location. This allows for storing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) for later use across various machine learning models.\n\nThe workspace is recommended to be on the same server as the implementation, this ensures better performance.",
     "categories": [
         "Export",
         "machine learning"
@@ -25,7 +25,7 @@
         },
         {
             "name": "folder",
-            "description": "The folder path within the server backend workspace where the data cube should be stored. This helps organize different data cubes within the workspace.",
+            "description": "The folder path within the workspace where the data cube should be stored.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals-ml/export_ml_model.json b/proposals-ml/export_ml_model.json
index 0c6cb781..cbf6fffa 100644
--- a/proposals-ml/export_ml_model.json
+++ b/proposals-ml/export_ml_model.json
@@ -1,7 +1,7 @@
 {
     "id": "export_ml_model",
     "summary": "Export a machine learning model to a workspace",
-    "description": "Exports a machine learning model to a specified workspace location. This is designed for intermediate model storage and can use the backend language's native serialization format.\n\nUnlike save_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for temporary storage and transfer of models between processes.",
+    "description": "Exports a machine learning model to a specified workspace location. This is designed for intermediate model storage and can use the implementation's native serialization format.\n\nUnlike save_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for temporary storage and transfer of models between processes.",
     "categories": [
         "Export",
         "machine learning"
@@ -10,7 +10,7 @@
     "parameters": [
         {
             "name": "model",
-            "description": "The machine learning model to be exported. This can be in any format supported by the backend implementation.",
+            "description": "The machine learning model to be exported. This can be in any format supported by the implementation.",
             "schema": {
                 "type": "object",
                 "subtype": "ml-model"
@@ -25,7 +25,7 @@
         },
         {
             "name": "folder",
-            "description": "The folder path within the server backend workspace where the model should be stored.",
+            "description": "The folder path within the workspace where the model should be stored.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals-ml/import_cube.json b/proposals-ml/import_cube.json
index b4d9987a..1591b65b 100644
--- a/proposals-ml/import_cube.json
+++ b/proposals-ml/import_cube.json
@@ -17,7 +17,7 @@
         },
         {
             "name": "folder",
-            "description": "The folder path within the server backend workspace where the data cube is stored. This should match the folder path used when the data cube was exported.",
+            "description": "The folder path within the workspace where the data cube is stored.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals-ml/import_ml_model.json b/proposals-ml/import_ml_model.json
index ede1888b..5ba57f61 100644
--- a/proposals-ml/import_ml_model.json
+++ b/proposals-ml/import_ml_model.json
@@ -1,7 +1,7 @@
 {
     "id": "import_ml_model",
     "summary": "Import a machine learning model from a workspace",
-    "description": "Imports a previously exported machine learning model from a specified workspace location. This is designed for intermediate model retrieval and can handle models in the backend language's native serialization format.\n\nUnlike load_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for retrieving temporarily stored models between processes.",
+    "description": "Imports a previously exported machine learning model from a specified workspace location. This is designed for intermediate model retrieval and can handle models in the implementation's native serialization format.\n\nUnlike load_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for retrieving temporarily stored models between processes.",
     "categories": [
         "Import",
         "machine learning"
@@ -17,7 +17,7 @@
         },
         {
             "name": "folder",
-            "description": "The folder path within the server backend workspace where the model is stored.",
+            "description": "The folder path within the workspace where the model is stored.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals/export_cube.json b/proposals/export_cube.json
index a1292b0e..8810c653 100644
--- a/proposals/export_cube.json
+++ b/proposals/export_cube.json
@@ -1,7 +1,7 @@
 {
     "id": "export_cube",
     "summary": "Export a data cube to a workspace",
-    "description": "Exports a processed data cube to a specified workspace location. This allows for storing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) for later use across various machine learning models.\n\nThe workspace is recommended to be on the server backend,this ensures better performance.",
+    "description": "Exports a processed data cube to a specified workspace location. This allows for storing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) for later use across various machine learning models.\n\nThe workspace is recommended to be on the same server as the implementation, this ensures better performance.",
     "categories": [
         "Export",
         "machine learning"
@@ -25,7 +25,7 @@
         },
         {
             "name": "folder",
-            "description": "The folder path within the server backend workspace where the data cube should be stored. This helps organize different data cubes within the workspace.",
+            "description": "The folder path within the workspace where the data cube should be stored.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals/export_ml_model.json b/proposals/export_ml_model.json
index 0c6cb781..cbf6fffa 100644
--- a/proposals/export_ml_model.json
+++ b/proposals/export_ml_model.json
@@ -1,7 +1,7 @@
 {
     "id": "export_ml_model",
     "summary": "Export a machine learning model to a workspace",
-    "description": "Exports a machine learning model to a specified workspace location. This is designed for intermediate model storage and can use the backend language's native serialization format.\n\nUnlike save_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for temporary storage and transfer of models between processes.",
+    "description": "Exports a machine learning model to a specified workspace location. This is designed for intermediate model storage and can use the implementation's native serialization format.\n\nUnlike save_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for temporary storage and transfer of models between processes.",
     "categories": [
         "Export",
         "machine learning"
@@ -10,7 +10,7 @@
     "parameters": [
         {
             "name": "model",
-            "description": "The machine learning model to be exported. This can be in any format supported by the backend implementation.",
+            "description": "The machine learning model to be exported. This can be in any format supported by the implementation.",
             "schema": {
                 "type": "object",
                 "subtype": "ml-model"
@@ -25,7 +25,7 @@
         },
         {
             "name": "folder",
-            "description": "The folder path within the server backend workspace where the model should be stored.",
+            "description": "The folder path within the workspace where the model should be stored.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals/import_cube.json b/proposals/import_cube.json
index b4d9987a..1591b65b 100644
--- a/proposals/import_cube.json
+++ b/proposals/import_cube.json
@@ -17,7 +17,7 @@
         },
         {
             "name": "folder",
-            "description": "The folder path within the server backend workspace where the data cube is stored. This should match the folder path used when the data cube was exported.",
+            "description": "The folder path within the workspace where the data cube is stored.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals/import_ml_model.json b/proposals/import_ml_model.json
index ede1888b..5ba57f61 100644
--- a/proposals/import_ml_model.json
+++ b/proposals/import_ml_model.json
@@ -1,7 +1,7 @@
 {
     "id": "import_ml_model",
     "summary": "Import a machine learning model from a workspace",
-    "description": "Imports a previously exported machine learning model from a specified workspace location. This is designed for intermediate model retrieval and can handle models in the backend language's native serialization format.\n\nUnlike load_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for retrieving temporarily stored models between processes.",
+    "description": "Imports a previously exported machine learning model from a specified workspace location. This is designed for intermediate model retrieval and can handle models in the implementation's native serialization format.\n\nUnlike load_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for retrieving temporarily stored models between processes.",
     "categories": [
         "Import",
         "machine learning"
@@ -17,7 +17,7 @@
         },
         {
             "name": "folder",
-            "description": "The folder path within the server backend workspace where the model is stored.",
+            "description": "The folder path within the workspace where the model is stored.",
             "schema": {
                 "type": "string"
             }

From 04d314a990306bc2b514feaa71c2c0907c1958a9 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Wed, 16 Apr 2025 23:43:04 +0200
Subject: [PATCH 18/52] ml uncertainty

---
 CHANGELOG.md                           |  1 +
 proposals-ml/ml_uncertainty_class.json | 40 ++++++++++++++++++++++++++
 proposals/ml_uncertainty_class.json    | 40 ++++++++++++++++++++++++++
 3 files changed, 81 insertions(+)
 create mode 100644 proposals-ml/ml_uncertainty_class.json
 create mode 100644 proposals/ml_uncertainty_class.json

diff --git a/CHANGELOG.md b/CHANGELOG.md
index bffcc5fc..b47b23a3 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -33,6 +33,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `ml_fit`
     - `ml_predict`
     - `ml_predict_probabilities`
+    - `ml_uncertainty_class`
     - `save_ml_model`
     - `unflatten_dimension`
     - `vector_buffer`
diff --git a/proposals-ml/ml_uncertainty_class.json b/proposals-ml/ml_uncertainty_class.json
new file mode 100644
index 00000000..7c98bbfc
--- /dev/null
+++ b/proposals-ml/ml_uncertainty_class.json
@@ -0,0 +1,40 @@
+{
+    "id": "ml_uncertainty_class",
+    "summary": "Estimate uncertainty of classification probabilities",
+    "description": "Calculates uncertainty estimates for classification probabilities using various methods. The uncertainty is derived from the probability distribution of class predictions.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The data cube containing the classification probabilities.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "method",
+            "description": "The method to use for uncertainty estimation. 'margin' calculates the difference between the two highest probabilities, 'ratio' uses the ratio of the two highest probabilities, and 'least_confidence' uses 1 minus the highest probability.",
+            "default": "margin",
+            "optional": true,
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "margin",
+                    "ratio",
+                    "least_confidence"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A data cube with the uncertainty estimates. The output maintains the same dimensions as the input data cube, but replaces the class probabilities with uncertainty values. Higher values indicate higher uncertainty in the classification.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/proposals/ml_uncertainty_class.json b/proposals/ml_uncertainty_class.json
new file mode 100644
index 00000000..7c98bbfc
--- /dev/null
+++ b/proposals/ml_uncertainty_class.json
@@ -0,0 +1,40 @@
+{
+    "id": "ml_uncertainty_class",
+    "summary": "Estimate uncertainty of classification probabilities",
+    "description": "Calculates uncertainty estimates for classification probabilities using various methods. The uncertainty is derived from the probability distribution of class predictions.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The data cube containing the classification probabilities.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "method",
+            "description": "The method to use for uncertainty estimation. 'margin' calculates the difference between the two highest probabilities, 'ratio' uses the ratio of the two highest probabilities, and 'least_confidence' uses 1 minus the highest probability.",
+            "default": "margin",
+            "optional": true,
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "margin",
+                    "ratio",
+                    "least_confidence"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A data cube with the uncertainty estimates. The output maintains the same dimensions as the input data cube, but replaces the class probabilities with uncertainty values. Higher values indicate higher uncertainty in the classification.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    }
+} 
\ No newline at end of file

From 58768cc20883cd2c7e49c357d29ff7d7d3e59580 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Wed, 16 Apr 2025 23:47:51 +0200
Subject: [PATCH 19/52] least-confidence

---
 proposals-ml/ml_uncertainty_class.json | 4 ++--
 proposals/ml_uncertainty_class.json    | 4 ++--
 tests/.words                           | 1 +
 3 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/proposals-ml/ml_uncertainty_class.json b/proposals-ml/ml_uncertainty_class.json
index 7c98bbfc..86402906 100644
--- a/proposals-ml/ml_uncertainty_class.json
+++ b/proposals-ml/ml_uncertainty_class.json
@@ -17,7 +17,7 @@
         },
         {
             "name": "method",
-            "description": "The method to use for uncertainty estimation. 'margin' calculates the difference between the two highest probabilities, 'ratio' uses the ratio of the two highest probabilities, and 'least_confidence' uses 1 minus the highest probability.",
+            "description": "The method to use for uncertainty estimation. 'margin' calculates the difference between the two highest probabilities, 'ratio' uses the ratio of the two highest probabilities, and 'least-confidence' uses 1 minus the highest probability.",
             "default": "margin",
             "optional": true,
             "schema": {
@@ -25,7 +25,7 @@
                 "enum": [
                     "margin",
                     "ratio",
-                    "least_confidence"
+                    "least-confidence"
                 ]
             }
         }
diff --git a/proposals/ml_uncertainty_class.json b/proposals/ml_uncertainty_class.json
index 7c98bbfc..86402906 100644
--- a/proposals/ml_uncertainty_class.json
+++ b/proposals/ml_uncertainty_class.json
@@ -17,7 +17,7 @@
         },
         {
             "name": "method",
-            "description": "The method to use for uncertainty estimation. 'margin' calculates the difference between the two highest probabilities, 'ratio' uses the ratio of the two highest probabilities, and 'least_confidence' uses 1 minus the highest probability.",
+            "description": "The method to use for uncertainty estimation. 'margin' calculates the difference between the two highest probabilities, 'ratio' uses the ratio of the two highest probabilities, and 'least-confidence' uses 1 minus the highest probability.",
             "default": "margin",
             "optional": true,
             "schema": {
@@ -25,7 +25,7 @@
                 "enum": [
                     "margin",
                     "ratio",
-                    "least_confidence"
+                    "least-confidence"
                 ]
             }
         }
diff --git a/tests/.words b/tests/.words
index 8e601773..3e2ed97c 100644
--- a/tests/.words
+++ b/tests/.words
@@ -76,3 +76,4 @@ Giordano
 Chehata
 LightTAE
 TAE
+least-confidence

From 026db1b283592066786e70b65b10d3da0d956381 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Tue, 22 Apr 2025 17:40:25 +0200
Subject: [PATCH 20/52] ml classification smoothing

---
 .github/workflows/docs.yml        |  1 +
 CHANGELOG.md                      |  1 +
 proposals-ml/ml_smooth_class.json | 66 +++++++++++++++++++++++++++++++
 proposals/ml_smooth_class.json    | 66 +++++++++++++++++++++++++++++++
 tests/.words                      |  1 +
 5 files changed, 135 insertions(+)
 create mode 100644 proposals-ml/ml_smooth_class.json
 create mode 100644 proposals/ml_smooth_class.json

diff --git a/.github/workflows/docs.yml b/.github/workflows/docs.yml
index aff34e08..c8642243 100644
--- a/.github/workflows/docs.yml
+++ b/.github/workflows/docs.yml
@@ -5,6 +5,7 @@ on:
   push:
     branches:
       - dev
+      - ml-poc
       - draft
       - master
 jobs:
diff --git a/CHANGELOG.md b/CHANGELOG.md
index b47b23a3..7f5e16e0 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -33,6 +33,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `ml_fit`
     - `ml_predict`
     - `ml_predict_probabilities`
+    - `ml_smooth_class`
     - `ml_uncertainty_class`
     - `save_ml_model`
     - `unflatten_dimension`
diff --git a/proposals-ml/ml_smooth_class.json b/proposals-ml/ml_smooth_class.json
new file mode 100644
index 00000000..16499608
--- /dev/null
+++ b/proposals-ml/ml_smooth_class.json
@@ -0,0 +1,66 @@
+{
+    "id": "ml_smooth_class",
+    "summary": "Apply smoothing to classification probabilities results using Bayesian inference",
+    "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighborhood information. The smoothing is based on a window-based approach.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The input classification probabilities data to be smoothed. This should be a classified datacube.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "window_size",
+            "description": "The size of the moving window used for spatial smoothing. Must be an odd number to ensure a central pixel.",
+            "optional": true,
+            "default": 7,
+            "schema": {
+                "type": "integer",
+                "minimum": 3,
+                "multipleOf": 2
+            }
+        },
+        {
+            "name": "neighborhood_fraction",
+            "description": "The fraction of neighboring pixels to consider in the smoothing process. Controls the influence of local spatial information.",
+            "optional": true,
+            "default": 0.5,
+            "schema": {
+                "type": "number",
+                "minimum": 0,
+                "maximum": 1
+            }
+        },
+        {
+            "name": "smoothness",
+            "description": "Controls the degree of smoothing applied. Higher values result in smoother outputs but may lose fine details.",
+            "optional": true,
+            "default": 10,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        }
+    ],
+    "returns": {
+        "description": "A smoothed classification result with reduced noise and improved spatial consistency.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.3390/rs16234572",
+            "title": "Camara et al. (2024): Bayesian Inference for Post-Processing of Remote-Sensing Image Classification",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals/ml_smooth_class.json b/proposals/ml_smooth_class.json
new file mode 100644
index 00000000..16499608
--- /dev/null
+++ b/proposals/ml_smooth_class.json
@@ -0,0 +1,66 @@
+{
+    "id": "ml_smooth_class",
+    "summary": "Apply smoothing to classification probabilities results using Bayesian inference",
+    "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighborhood information. The smoothing is based on a window-based approach.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The input classification probabilities data to be smoothed. This should be a classified datacube.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "window_size",
+            "description": "The size of the moving window used for spatial smoothing. Must be an odd number to ensure a central pixel.",
+            "optional": true,
+            "default": 7,
+            "schema": {
+                "type": "integer",
+                "minimum": 3,
+                "multipleOf": 2
+            }
+        },
+        {
+            "name": "neighborhood_fraction",
+            "description": "The fraction of neighboring pixels to consider in the smoothing process. Controls the influence of local spatial information.",
+            "optional": true,
+            "default": 0.5,
+            "schema": {
+                "type": "number",
+                "minimum": 0,
+                "maximum": 1
+            }
+        },
+        {
+            "name": "smoothness",
+            "description": "Controls the degree of smoothing applied. Higher values result in smoother outputs but may lose fine details.",
+            "optional": true,
+            "default": 10,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        }
+    ],
+    "returns": {
+        "description": "A smoothed classification result with reduced noise and improved spatial consistency.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.3390/rs16234572",
+            "title": "Camara et al. (2024): Bayesian Inference for Post-Processing of Remote-Sensing Image Classification",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/tests/.words b/tests/.words
index 3e2ed97c..573e87c5 100644
--- a/tests/.words
+++ b/tests/.words
@@ -77,3 +77,4 @@ Chehata
 LightTAE
 TAE
 least-confidence
+Camara

From 62c1d84225cb8da8956253ec2903cdbde58c6cc0 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Tue, 22 Apr 2025 22:26:54 +0200
Subject: [PATCH 21/52] smooth refactor

---
 proposals-ml/ml_smooth_class.json | 8 ++++----
 proposals/ml_smooth_class.json    | 8 ++++----
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/proposals-ml/ml_smooth_class.json b/proposals-ml/ml_smooth_class.json
index 16499608..0efd7aa5 100644
--- a/proposals-ml/ml_smooth_class.json
+++ b/proposals-ml/ml_smooth_class.json
@@ -1,7 +1,7 @@
 {
     "id": "ml_smooth_class",
-    "summary": "Apply smoothing to classification probabilities results using Bayesian inference",
-    "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighborhood information. The smoothing is based on a window-based approach.",
+    "summary": "Apply Bayesian smoothing to classification probability results",
+    "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighbourhood information. The smoothing is based on a window-based approach.",
     "categories": [
         "machine learning"
     ],
@@ -9,7 +9,7 @@
     "parameters": [
         {
             "name": "data",
-            "description": "The input classification probabilities data to be smoothed. This should be a classified datacube.",
+            "description": "The input classification probabilities data to be smoothed. This should be a classified data cube.",
             "schema": {
                 "type": "object",
                 "subtype": "datacube"
@@ -28,7 +28,7 @@
         },
         {
             "name": "neighborhood_fraction",
-            "description": "The fraction of neighboring pixels to consider in the smoothing process. Controls the influence of local spatial information.",
+            "description": "The fraction of neighbouring pixels to consider in the smoothing process. Controls the influence of local spatial information.",
             "optional": true,
             "default": 0.5,
             "schema": {
diff --git a/proposals/ml_smooth_class.json b/proposals/ml_smooth_class.json
index 16499608..0efd7aa5 100644
--- a/proposals/ml_smooth_class.json
+++ b/proposals/ml_smooth_class.json
@@ -1,7 +1,7 @@
 {
     "id": "ml_smooth_class",
-    "summary": "Apply smoothing to classification probabilities results using Bayesian inference",
-    "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighborhood information. The smoothing is based on a window-based approach.",
+    "summary": "Apply Bayesian smoothing to classification probability results",
+    "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighbourhood information. The smoothing is based on a window-based approach.",
     "categories": [
         "machine learning"
     ],
@@ -9,7 +9,7 @@
     "parameters": [
         {
             "name": "data",
-            "description": "The input classification probabilities data to be smoothed. This should be a classified datacube.",
+            "description": "The input classification probabilities data to be smoothed. This should be a classified data cube.",
             "schema": {
                 "type": "object",
                 "subtype": "datacube"
@@ -28,7 +28,7 @@
         },
         {
             "name": "neighborhood_fraction",
-            "description": "The fraction of neighboring pixels to consider in the smoothing process. Controls the influence of local spatial information.",
+            "description": "The fraction of neighbouring pixels to consider in the smoothing process. Controls the influence of local spatial information.",
             "optional": true,
             "default": 0.5,
             "schema": {

From d9bf489eb9361c33631595278211e347a5604f6f Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Tue, 22 Apr 2025 22:31:00 +0200
Subject: [PATCH 22/52] smooth shorten summary

---
 proposals-ml/ml_smooth_class.json | 2 +-
 proposals/ml_smooth_class.json    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/proposals-ml/ml_smooth_class.json b/proposals-ml/ml_smooth_class.json
index 0efd7aa5..380de6ad 100644
--- a/proposals-ml/ml_smooth_class.json
+++ b/proposals-ml/ml_smooth_class.json
@@ -1,6 +1,6 @@
 {
     "id": "ml_smooth_class",
-    "summary": "Apply Bayesian smoothing to classification probability results",
+    "summary": "Apply Bayesian smoothing to classification results",
     "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighbourhood information. The smoothing is based on a window-based approach.",
     "categories": [
         "machine learning"
diff --git a/proposals/ml_smooth_class.json b/proposals/ml_smooth_class.json
index 0efd7aa5..380de6ad 100644
--- a/proposals/ml_smooth_class.json
+++ b/proposals/ml_smooth_class.json
@@ -1,6 +1,6 @@
 {
     "id": "ml_smooth_class",
-    "summary": "Apply Bayesian smoothing to classification probability results",
+    "summary": "Apply Bayesian smoothing to classification results",
     "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighbourhood information. The smoothing is based on a window-based approach.",
     "categories": [
         "machine learning"

From 55b108bb7a2875e0746c40ce511b37bd546a4a3a Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Wed, 23 Apr 2025 08:25:15 +0200
Subject: [PATCH 23/52] ml smooth classification (#8)

* ml classification smoothing

* smooth refactor

* smooth shorten summary
---
 .github/workflows/docs.yml        |  1 +
 CHANGELOG.md                      |  1 +
 proposals-ml/ml_smooth_class.json | 66 +++++++++++++++++++++++++++++++
 proposals/ml_smooth_class.json    | 66 +++++++++++++++++++++++++++++++
 tests/.words                      |  1 +
 5 files changed, 135 insertions(+)
 create mode 100644 proposals-ml/ml_smooth_class.json
 create mode 100644 proposals/ml_smooth_class.json

diff --git a/.github/workflows/docs.yml b/.github/workflows/docs.yml
index aff34e08..c8642243 100644
--- a/.github/workflows/docs.yml
+++ b/.github/workflows/docs.yml
@@ -5,6 +5,7 @@ on:
   push:
     branches:
       - dev
+      - ml-poc
       - draft
       - master
 jobs:
diff --git a/CHANGELOG.md b/CHANGELOG.md
index b47b23a3..7f5e16e0 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -33,6 +33,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `ml_fit`
     - `ml_predict`
     - `ml_predict_probabilities`
+    - `ml_smooth_class`
     - `ml_uncertainty_class`
     - `save_ml_model`
     - `unflatten_dimension`
diff --git a/proposals-ml/ml_smooth_class.json b/proposals-ml/ml_smooth_class.json
new file mode 100644
index 00000000..380de6ad
--- /dev/null
+++ b/proposals-ml/ml_smooth_class.json
@@ -0,0 +1,66 @@
+{
+    "id": "ml_smooth_class",
+    "summary": "Apply Bayesian smoothing to classification results",
+    "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighbourhood information. The smoothing is based on a window-based approach.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The input classification probabilities data to be smoothed. This should be a classified data cube.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "window_size",
+            "description": "The size of the moving window used for spatial smoothing. Must be an odd number to ensure a central pixel.",
+            "optional": true,
+            "default": 7,
+            "schema": {
+                "type": "integer",
+                "minimum": 3,
+                "multipleOf": 2
+            }
+        },
+        {
+            "name": "neighborhood_fraction",
+            "description": "The fraction of neighbouring pixels to consider in the smoothing process. Controls the influence of local spatial information.",
+            "optional": true,
+            "default": 0.5,
+            "schema": {
+                "type": "number",
+                "minimum": 0,
+                "maximum": 1
+            }
+        },
+        {
+            "name": "smoothness",
+            "description": "Controls the degree of smoothing applied. Higher values result in smoother outputs but may lose fine details.",
+            "optional": true,
+            "default": 10,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        }
+    ],
+    "returns": {
+        "description": "A smoothed classification result with reduced noise and improved spatial consistency.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.3390/rs16234572",
+            "title": "Camara et al. (2024): Bayesian Inference for Post-Processing of Remote-Sensing Image Classification",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals/ml_smooth_class.json b/proposals/ml_smooth_class.json
new file mode 100644
index 00000000..380de6ad
--- /dev/null
+++ b/proposals/ml_smooth_class.json
@@ -0,0 +1,66 @@
+{
+    "id": "ml_smooth_class",
+    "summary": "Apply Bayesian smoothing to classification results",
+    "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighbourhood information. The smoothing is based on a window-based approach.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The input classification probabilities data to be smoothed. This should be a classified data cube.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "window_size",
+            "description": "The size of the moving window used for spatial smoothing. Must be an odd number to ensure a central pixel.",
+            "optional": true,
+            "default": 7,
+            "schema": {
+                "type": "integer",
+                "minimum": 3,
+                "multipleOf": 2
+            }
+        },
+        {
+            "name": "neighborhood_fraction",
+            "description": "The fraction of neighbouring pixels to consider in the smoothing process. Controls the influence of local spatial information.",
+            "optional": true,
+            "default": 0.5,
+            "schema": {
+                "type": "number",
+                "minimum": 0,
+                "maximum": 1
+            }
+        },
+        {
+            "name": "smoothness",
+            "description": "Controls the degree of smoothing applied. Higher values result in smoother outputs but may lose fine details.",
+            "optional": true,
+            "default": 10,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        }
+    ],
+    "returns": {
+        "description": "A smoothed classification result with reduced noise and improved spatial consistency.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.3390/rs16234572",
+            "title": "Camara et al. (2024): Bayesian Inference for Post-Processing of Remote-Sensing Image Classification",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/tests/.words b/tests/.words
index 3e2ed97c..573e87c5 100644
--- a/tests/.words
+++ b/tests/.words
@@ -77,3 +77,4 @@ Chehata
 LightTAE
 TAE
 least-confidence
+Camara

From 111c023a95e86e8008eefd92520e7ba942a7173e Mon Sep 17 00:00:00 2001
From: Jonas Hurst <76062450+jonas-hurst@users.noreply.github.com>
Date: Wed, 23 Apr 2025 08:27:53 +0200
Subject: [PATCH 24/52] altered load_ml_model process spec (#6)

* added detail to description

* added asset_name parameter

* alter schema

* mirror spec from proposals-ml folder

* added hyperlink to mlm
---
 proposals-ml/load_ml_model.json | 11 ++++++++++-
 proposals/load_ml_model.json    | 11 ++++++++++-
 2 files changed, 20 insertions(+), 2 deletions(-)

diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_ml_model.json
index 4b4e9ed4..bbc69d4b 100644
--- a/proposals-ml/load_ml_model.json
+++ b/proposals-ml/load_ml_model.json
@@ -10,7 +10,7 @@
     "parameters": [
         {
             "name": "uri",
-            "description": "The STAC Item to load the machine learning model from. The STAC Item must implement the `mlm-model` extension.",
+            "description": "The STAC Item to load the machine learning model from. The STAC Item must implement the [`mlm`](https://github.com/stac-extensions/mlm) extension. This parameter can point to a remote STAC Item via ``URL`` or a local JSON file.",
             "schema": [
                 {
                     "title": "URL",
@@ -26,6 +26,15 @@
                     "pattern": "^[^\r\n\\:'\"]+$"
                 }
             ]
+        },
+        {
+            "name": "model_asset",
+            "description": "The Asset name of the given STAC Item which represents the actual ML model. The asset must list ``mlm:model`` as its role. If only one asset lists ``mlm:model`` as its role, this parameter is optional as this asset will be used by default. If multiple assets list ``mlm:model`` as their role, this parameter is required to determine which asset to use.",
+            "schema": {
+                "type": "string"
+            },
+            "default": null,
+            "optinoal": true
         }
     ],
     "returns": {
diff --git a/proposals/load_ml_model.json b/proposals/load_ml_model.json
index 4b4e9ed4..bbc69d4b 100644
--- a/proposals/load_ml_model.json
+++ b/proposals/load_ml_model.json
@@ -10,7 +10,7 @@
     "parameters": [
         {
             "name": "uri",
-            "description": "The STAC Item to load the machine learning model from. The STAC Item must implement the `mlm-model` extension.",
+            "description": "The STAC Item to load the machine learning model from. The STAC Item must implement the [`mlm`](https://github.com/stac-extensions/mlm) extension. This parameter can point to a remote STAC Item via ``URL`` or a local JSON file.",
             "schema": [
                 {
                     "title": "URL",
@@ -26,6 +26,15 @@
                     "pattern": "^[^\r\n\\:'\"]+$"
                 }
             ]
+        },
+        {
+            "name": "model_asset",
+            "description": "The Asset name of the given STAC Item which represents the actual ML model. The asset must list ``mlm:model`` as its role. If only one asset lists ``mlm:model`` as its role, this parameter is optional as this asset will be used by default. If multiple assets list ``mlm:model`` as their role, this parameter is required to determine which asset to use.",
+            "schema": {
+                "type": "string"
+            },
+            "default": null,
+            "optinoal": true
         }
     ],
     "returns": {

From cf9d4d42348b0e4942bd2d45fcb336cb49f8da30 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Wed, 23 Apr 2025 15:20:39 +0200
Subject: [PATCH 25/52] ml label class

---
 CHANGELOG.md                     |  1 +
 proposals-ml/ml_label_class.json | 38 ++++++++++++++++++++++++++++++++
 proposals/ml_label_class.json    | 38 ++++++++++++++++++++++++++++++++
 3 files changed, 77 insertions(+)
 create mode 100644 proposals-ml/ml_label_class.json
 create mode 100644 proposals/ml_label_class.json

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 7f5e16e0..4562396f 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -31,6 +31,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `mlm_class_xgboost`
     - `mlm_regr_random_forest`
     - `ml_fit`
+    - `ml_label_class`
     - `ml_predict`
     - `ml_predict_probabilities`
     - `ml_smooth_class`
diff --git a/proposals-ml/ml_label_class.json b/proposals-ml/ml_label_class.json
new file mode 100644
index 00000000..649fe0e2
--- /dev/null
+++ b/proposals-ml/ml_label_class.json
@@ -0,0 +1,38 @@
+{
+    "id": "ml_label_class",
+    "summary": "Convert probability data cube to labeled data cube",
+    "description": "Converts a probability data cube to a labeled data cube by applying softmax normalization and selecting the class with the highest probability. Optionally allows mapping of class indices to custom labels.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The input probability data cube to be labeled. Each band should represent the probability of a different class.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "labels",
+            "description": "Optional dictionary mapping class indices to custom labels. The dictionary keys are the class indices (as integers) and the values are the custom labels for each class. If not provided, the class indices will be used as labels.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "object",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A labeled data cube where each pixel contains the class label with the highest probability after softmax normalization.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/proposals/ml_label_class.json b/proposals/ml_label_class.json
new file mode 100644
index 00000000..649fe0e2
--- /dev/null
+++ b/proposals/ml_label_class.json
@@ -0,0 +1,38 @@
+{
+    "id": "ml_label_class",
+    "summary": "Convert probability data cube to labeled data cube",
+    "description": "Converts a probability data cube to a labeled data cube by applying softmax normalization and selecting the class with the highest probability. Optionally allows mapping of class indices to custom labels.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The input probability data cube to be labeled. Each band should represent the probability of a different class.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "labels",
+            "description": "Optional dictionary mapping class indices to custom labels. The dictionary keys are the class indices (as integers) and the values are the custom labels for each class. If not provided, the class indices will be used as labels.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "object",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A labeled data cube where each pixel contains the class label with the highest probability after softmax normalization.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    }
+} 
\ No newline at end of file

From 85345c6531931526d9234e2f242e35d1f136f988 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Wed, 23 Apr 2025 15:43:23 +0200
Subject: [PATCH 26/52] softmax word

---
 tests/.words | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/.words b/tests/.words
index 573e87c5..fb40804c 100644
--- a/tests/.words
+++ b/tests/.words
@@ -78,3 +78,4 @@ LightTAE
 TAE
 least-confidence
 Camara
+softmax

From 504543c8255651e193bc39c7a7d1a7d1ce237d6f Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Wed, 23 Apr 2025 15:48:50 +0200
Subject: [PATCH 27/52] ml label class (#9)

* ml classification smoothing

* smooth refactor

* smooth shorten summary

* ml label class

* softmax word
---
 CHANGELOG.md                     |  1 +
 proposals-ml/ml_label_class.json | 38 ++++++++++++++++++++++++++++++++
 proposals/ml_label_class.json    | 38 ++++++++++++++++++++++++++++++++
 tests/.words                     |  1 +
 4 files changed, 78 insertions(+)
 create mode 100644 proposals-ml/ml_label_class.json
 create mode 100644 proposals/ml_label_class.json

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 7f5e16e0..4562396f 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -31,6 +31,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `mlm_class_xgboost`
     - `mlm_regr_random_forest`
     - `ml_fit`
+    - `ml_label_class`
     - `ml_predict`
     - `ml_predict_probabilities`
     - `ml_smooth_class`
diff --git a/proposals-ml/ml_label_class.json b/proposals-ml/ml_label_class.json
new file mode 100644
index 00000000..649fe0e2
--- /dev/null
+++ b/proposals-ml/ml_label_class.json
@@ -0,0 +1,38 @@
+{
+    "id": "ml_label_class",
+    "summary": "Convert probability data cube to labeled data cube",
+    "description": "Converts a probability data cube to a labeled data cube by applying softmax normalization and selecting the class with the highest probability. Optionally allows mapping of class indices to custom labels.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The input probability data cube to be labeled. Each band should represent the probability of a different class.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "labels",
+            "description": "Optional dictionary mapping class indices to custom labels. The dictionary keys are the class indices (as integers) and the values are the custom labels for each class. If not provided, the class indices will be used as labels.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "object",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A labeled data cube where each pixel contains the class label with the highest probability after softmax normalization.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/proposals/ml_label_class.json b/proposals/ml_label_class.json
new file mode 100644
index 00000000..649fe0e2
--- /dev/null
+++ b/proposals/ml_label_class.json
@@ -0,0 +1,38 @@
+{
+    "id": "ml_label_class",
+    "summary": "Convert probability data cube to labeled data cube",
+    "description": "Converts a probability data cube to a labeled data cube by applying softmax normalization and selecting the class with the highest probability. Optionally allows mapping of class indices to custom labels.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "data",
+            "description": "The input probability data cube to be labeled. Each band should represent the probability of a different class.",
+            "schema": {
+                "type": "object",
+                "subtype": "datacube"
+            }
+        },
+        {
+            "name": "labels",
+            "description": "Optional dictionary mapping class indices to custom labels. The dictionary keys are the class indices (as integers) and the values are the custom labels for each class. If not provided, the class indices will be used as labels.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "object",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A labeled data cube where each pixel contains the class label with the highest probability after softmax normalization.",
+        "schema": {
+            "type": "object",
+            "subtype": "datacube"
+        }
+    }
+} 
\ No newline at end of file
diff --git a/tests/.words b/tests/.words
index 573e87c5..fb40804c 100644
--- a/tests/.words
+++ b/tests/.words
@@ -78,3 +78,4 @@ LightTAE
 TAE
 least-confidence
 Camara
+softmax

From 0191b1ed2d0a2e4569c4459f906c2b488591c960 Mon Sep 17 00:00:00 2001
From: Brian Pondi <brian.pondi@uni-muenster.de>
Date: Mon, 5 May 2025 22:45:16 +0200
Subject: [PATCH 28/52] 1d cnn

---
 proposals-ml/mlm_class_1dcnn.json | 178 ++++++++++++++++++++++++++++++
 1 file changed, 178 insertions(+)
 create mode 100644 proposals-ml/mlm_class_1dcnn.json

diff --git a/proposals-ml/mlm_class_1dcnn.json b/proposals-ml/mlm_class_1dcnn.json
new file mode 100644
index 00000000..887276c0
--- /dev/null
+++ b/proposals-ml/mlm_class_1dcnn.json
@@ -0,0 +1,178 @@
+{
+    "id": "mlm_class_1dcnn",
+    "summary": "Initialize a 1D CNN classification model",
+    "description": "Initializes a 1D Convolutional Neural Network (CNN) classification model. The number of input channels and output classes are inferred automatically from the training data at fit time. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "conv_filters",
+            "description": "List of integers specifying the number of filters in each convolutional layer. The final output layer for classification will be added automatically based on the number of classes in the training data.",
+            "default": [64, 128, 256, 512],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "conv_kernels",
+            "description": "List of integers or 'global' for kernel size in each convolutional layer. The 4th conv uses a global kernel (covers the entire sequence at that layer). The final output layer for classification will use a 1x1 kernel and be added automatically.",
+            "default": [3, 3, 3, "global"],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "oneOf": [
+                        {"type": "integer", "minimum": 1},
+                        {"type": "string", "enum": ["global"]}
+                    ]
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "conv_strides",
+            "description": "List of integers specifying the stride for each convolutional layer. The final output layer for classification will use stride 1 and be added automatically.",
+            "default": [1, 1, 1, 1],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "use_batchnorm",
+            "description": "List of booleans specifying whether to use batch normalization after each convolutional layer. The final output layer will not use batch normalization.",
+            "default": [true, true, true, false],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "boolean"
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "activation",
+            "description": "Activation function to use after each convolutional layer.",
+            "default": "relu",
+            "schema": {
+                "type": "string",
+                "enum": ["relu", "tanh", "sigmoid", "leakyrelu"]
+            }
+        },
+        {
+            "name": "maxpool_sizes",
+            "description": "List of integers specifying the pool size for each max pooling layer (after each of the first 3 conv blocks).",
+            "default": [2, 2, 2],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 3,
+                "maxItems": 3
+            }
+        },
+        {
+            "name": "maxpool_strides",
+            "description": "List of integers specifying the stride for each max pooling layer.",
+            "default": [2, 2, 2],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 3,
+                "maxItems": 3
+            }
+        },
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 100,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 64,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "adam",
+                    "sgd",
+                    "rmsprop",
+                    "adagrad",
+                    "nadam"
+                ]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://ieeexplore.ieee.org/document/8921180",
+            "title": "Song et al. (2019): Land Cover Classification for Satellite Images Through 1D CNN",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file

From d3762eb20acb902674414fadde15479513d3b670 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Wed, 28 May 2025 11:21:56 +0200
Subject: [PATCH 29/52] svr and catboost class

---
 CHANGELOG.md                         |   2 +
 proposals-ml/mlm_class_catboost.json |  62 ++++++++++++++
 proposals-ml/mlm_regr_svm.json       | 122 +++++++++++++++++++++++++++
 proposals/mlm_class_catboost.json    |  62 ++++++++++++++
 proposals/mlm_regr_svm.json          | 122 +++++++++++++++++++++++++++
 tests/.words                         |   4 +
 6 files changed, 374 insertions(+)
 create mode 100644 proposals-ml/mlm_class_catboost.json
 create mode 100644 proposals-ml/mlm_regr_svm.json
 create mode 100644 proposals/mlm_class_catboost.json
 create mode 100644 proposals/mlm_regr_svm.json

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 4562396f..b07d8349 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -22,6 +22,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `load_geojson`
     - `load_ml_model`
     - `load_url`
+    - `mlm_class_catboost`
     - `mlm_class_lighttae`
     - `mlm_class_mlp`
     - `mlm_class_random_forest`
@@ -30,6 +31,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `mlm_class_tempcnn`
     - `mlm_class_xgboost`
     - `mlm_regr_random_forest`
+    - `mlm_regr_svm`
     - `ml_fit`
     - `ml_label_class`
     - `ml_predict`
diff --git a/proposals-ml/mlm_class_catboost.json b/proposals-ml/mlm_class_catboost.json
new file mode 100644
index 00000000..7a46d0a1
--- /dev/null
+++ b/proposals-ml/mlm_class_catboost.json
@@ -0,0 +1,62 @@
+{
+    "id": "mlm_class_catboost",
+    "summary": "Initialize a CatBoost classification model",
+    "description": "Initializes a CatBoost classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "iterations",
+            "description": "The maximum number of trees that can be built during the training process.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 1,
+                "maximum": 500
+            }
+        },
+        {
+            "name": "depth",
+            "description": "Depth of the trees in the CatBoost model.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 1,
+                "maximum": 16
+            }
+        },
+        {
+            "name": "seed",
+            "description": "The random seed used for training, for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": 0,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ],
+                "minimum": 0,
+                "maximum": 2147483647
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://catboost.ai/",
+            "title": "CatBoost Documentation",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals-ml/mlm_regr_svm.json b/proposals-ml/mlm_regr_svm.json
new file mode 100644
index 00000000..ba63930c
--- /dev/null
+++ b/proposals-ml/mlm_regr_svm.json
@@ -0,0 +1,122 @@
+{
+    "id": "mlm_regr_svm",
+    "summary": "Initialize an SVM regression model",
+    "description": "Initializes a Support Vector Machine (SVM) regression model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "kernel",
+            "description": "Specifies the kernel type to be used in the algorithm.",
+            "optional": true,
+            "default": "rbf",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "linear",
+                    "poly",
+                    "rbf",
+                    "sigmoid"
+                ]
+            }
+        },
+        {
+            "name": "C",
+            "description": "Regularization parameter. The strength of the regularization is inversely proportional to C. Must be strictly positive.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "epsilon",
+            "description": "Epsilon in the epsilon-SVR model. Specifies the epsilon-tube within which no penalty is associated in the training loss function with points predicted within a distance epsilon from the actual value.",
+            "optional": true,
+            "default": 0.1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "gamma",
+            "description": "Kernel coefficient for 'rbf', 'poly', and 'sigmoid'. Higher values lead to tighter fits.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "degree",
+            "description": "Degree of the polynomial kernel function (only relevant for 'poly' kernel).",
+            "optional": true,
+            "default": 3,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "coef0",
+            "description": "Independent term in the kernel function (only relevant for 'poly' and 'sigmoid' kernels).",
+            "optional": true,
+            "default": 0,
+            "schema": {
+                "type": "number"
+            }
+        },
+        {
+            "name": "tolerance",
+            "description": "Tolerance of termination criterion.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "cachesize",
+            "description": "Size of the kernel cache in MB.",
+            "optional": true,
+            "default": 1000,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://link.springer.com/chapter/10.1007/978-1-4302-5990-9_4",
+            "title": "Awad, M., Khanna, R., Awad, M., & Khanna, R. (2015). Support vector regression. Efficient learning machines: Theories, concepts, and applications for engineers and system designers.",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals/mlm_class_catboost.json b/proposals/mlm_class_catboost.json
new file mode 100644
index 00000000..7a46d0a1
--- /dev/null
+++ b/proposals/mlm_class_catboost.json
@@ -0,0 +1,62 @@
+{
+    "id": "mlm_class_catboost",
+    "summary": "Initialize a CatBoost classification model",
+    "description": "Initializes a CatBoost classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "iterations",
+            "description": "The maximum number of trees that can be built during the training process.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 1,
+                "maximum": 500
+            }
+        },
+        {
+            "name": "depth",
+            "description": "Depth of the trees in the CatBoost model.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 1,
+                "maximum": 16
+            }
+        },
+        {
+            "name": "seed",
+            "description": "The random seed used for training, for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": 0,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ],
+                "minimum": 0,
+                "maximum": 2147483647
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://catboost.ai/",
+            "title": "CatBoost Documentation",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals/mlm_regr_svm.json b/proposals/mlm_regr_svm.json
new file mode 100644
index 00000000..ba63930c
--- /dev/null
+++ b/proposals/mlm_regr_svm.json
@@ -0,0 +1,122 @@
+{
+    "id": "mlm_regr_svm",
+    "summary": "Initialize an SVM regression model",
+    "description": "Initializes a Support Vector Machine (SVM) regression model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "kernel",
+            "description": "Specifies the kernel type to be used in the algorithm.",
+            "optional": true,
+            "default": "rbf",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "linear",
+                    "poly",
+                    "rbf",
+                    "sigmoid"
+                ]
+            }
+        },
+        {
+            "name": "C",
+            "description": "Regularization parameter. The strength of the regularization is inversely proportional to C. Must be strictly positive.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "epsilon",
+            "description": "Epsilon in the epsilon-SVR model. Specifies the epsilon-tube within which no penalty is associated in the training loss function with points predicted within a distance epsilon from the actual value.",
+            "optional": true,
+            "default": 0.1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "gamma",
+            "description": "Kernel coefficient for 'rbf', 'poly', and 'sigmoid'. Higher values lead to tighter fits.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "degree",
+            "description": "Degree of the polynomial kernel function (only relevant for 'poly' kernel).",
+            "optional": true,
+            "default": 3,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "coef0",
+            "description": "Independent term in the kernel function (only relevant for 'poly' and 'sigmoid' kernels).",
+            "optional": true,
+            "default": 0,
+            "schema": {
+                "type": "number"
+            }
+        },
+        {
+            "name": "tolerance",
+            "description": "Tolerance of termination criterion.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "cachesize",
+            "description": "Size of the kernel cache in MB.",
+            "optional": true,
+            "default": 1000,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://link.springer.com/chapter/10.1007/978-1-4302-5990-9_4",
+            "title": "Awad, M., Khanna, R., Awad, M., & Khanna, R. (2015). Support vector regression. Efficient learning machines: Theories, concepts, and applications for engineers and system designers.",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/tests/.words b/tests/.words
index fb40804c..3e820da0 100644
--- a/tests/.words
+++ b/tests/.words
@@ -79,3 +79,7 @@ TAE
 least-confidence
 Camara
 softmax
+Khanna
+Awad
+CatBoost
+epsilon-SVR

From b183ccbd80ad3b20f39faff52aa54dc201e15307 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Wed, 28 May 2025 11:44:32 +0200
Subject: [PATCH 30/52] svr and catboost (#10)

* ml classification smoothing

* smooth refactor

* smooth shorten summary

* ml label class

* softmax word

* 1d cnn

* svr and catboost class
---
 CHANGELOG.md                         |   2 +
 proposals-ml/mlm_class_1dcnn.json    | 178 +++++++++++++++++++++++++++
 proposals-ml/mlm_class_catboost.json |  62 ++++++++++
 proposals-ml/mlm_regr_svm.json       | 122 ++++++++++++++++++
 proposals/mlm_class_catboost.json    |  62 ++++++++++
 proposals/mlm_regr_svm.json          | 122 ++++++++++++++++++
 tests/.words                         |   4 +
 7 files changed, 552 insertions(+)
 create mode 100644 proposals-ml/mlm_class_1dcnn.json
 create mode 100644 proposals-ml/mlm_class_catboost.json
 create mode 100644 proposals-ml/mlm_regr_svm.json
 create mode 100644 proposals/mlm_class_catboost.json
 create mode 100644 proposals/mlm_regr_svm.json

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 4562396f..b07d8349 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -22,6 +22,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `load_geojson`
     - `load_ml_model`
     - `load_url`
+    - `mlm_class_catboost`
     - `mlm_class_lighttae`
     - `mlm_class_mlp`
     - `mlm_class_random_forest`
@@ -30,6 +31,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
     - `mlm_class_tempcnn`
     - `mlm_class_xgboost`
     - `mlm_regr_random_forest`
+    - `mlm_regr_svm`
     - `ml_fit`
     - `ml_label_class`
     - `ml_predict`
diff --git a/proposals-ml/mlm_class_1dcnn.json b/proposals-ml/mlm_class_1dcnn.json
new file mode 100644
index 00000000..887276c0
--- /dev/null
+++ b/proposals-ml/mlm_class_1dcnn.json
@@ -0,0 +1,178 @@
+{
+    "id": "mlm_class_1dcnn",
+    "summary": "Initialize a 1D CNN classification model",
+    "description": "Initializes a 1D Convolutional Neural Network (CNN) classification model. The number of input channels and output classes are inferred automatically from the training data at fit time. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "conv_filters",
+            "description": "List of integers specifying the number of filters in each convolutional layer. The final output layer for classification will be added automatically based on the number of classes in the training data.",
+            "default": [64, 128, 256, 512],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "conv_kernels",
+            "description": "List of integers or 'global' for kernel size in each convolutional layer. The 4th conv uses a global kernel (covers the entire sequence at that layer). The final output layer for classification will use a 1x1 kernel and be added automatically.",
+            "default": [3, 3, 3, "global"],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "oneOf": [
+                        {"type": "integer", "minimum": 1},
+                        {"type": "string", "enum": ["global"]}
+                    ]
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "conv_strides",
+            "description": "List of integers specifying the stride for each convolutional layer. The final output layer for classification will use stride 1 and be added automatically.",
+            "default": [1, 1, 1, 1],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "use_batchnorm",
+            "description": "List of booleans specifying whether to use batch normalization after each convolutional layer. The final output layer will not use batch normalization.",
+            "default": [true, true, true, false],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "boolean"
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "activation",
+            "description": "Activation function to use after each convolutional layer.",
+            "default": "relu",
+            "schema": {
+                "type": "string",
+                "enum": ["relu", "tanh", "sigmoid", "leakyrelu"]
+            }
+        },
+        {
+            "name": "maxpool_sizes",
+            "description": "List of integers specifying the pool size for each max pooling layer (after each of the first 3 conv blocks).",
+            "default": [2, 2, 2],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 3,
+                "maxItems": 3
+            }
+        },
+        {
+            "name": "maxpool_strides",
+            "description": "List of integers specifying the stride for each max pooling layer.",
+            "default": [2, 2, 2],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 3,
+                "maxItems": 3
+            }
+        },
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 100,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 64,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "adam",
+                    "sgd",
+                    "rmsprop",
+                    "adagrad",
+                    "nadam"
+                ]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://ieeexplore.ieee.org/document/8921180",
+            "title": "Song et al. (2019): Land Cover Classification for Satellite Images Through 1D CNN",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_catboost.json b/proposals-ml/mlm_class_catboost.json
new file mode 100644
index 00000000..7a46d0a1
--- /dev/null
+++ b/proposals-ml/mlm_class_catboost.json
@@ -0,0 +1,62 @@
+{
+    "id": "mlm_class_catboost",
+    "summary": "Initialize a CatBoost classification model",
+    "description": "Initializes a CatBoost classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "iterations",
+            "description": "The maximum number of trees that can be built during the training process.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 1,
+                "maximum": 500
+            }
+        },
+        {
+            "name": "depth",
+            "description": "Depth of the trees in the CatBoost model.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 1,
+                "maximum": 16
+            }
+        },
+        {
+            "name": "seed",
+            "description": "The random seed used for training, for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": 0,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ],
+                "minimum": 0,
+                "maximum": 2147483647
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://catboost.ai/",
+            "title": "CatBoost Documentation",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals-ml/mlm_regr_svm.json b/proposals-ml/mlm_regr_svm.json
new file mode 100644
index 00000000..ba63930c
--- /dev/null
+++ b/proposals-ml/mlm_regr_svm.json
@@ -0,0 +1,122 @@
+{
+    "id": "mlm_regr_svm",
+    "summary": "Initialize an SVM regression model",
+    "description": "Initializes a Support Vector Machine (SVM) regression model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "kernel",
+            "description": "Specifies the kernel type to be used in the algorithm.",
+            "optional": true,
+            "default": "rbf",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "linear",
+                    "poly",
+                    "rbf",
+                    "sigmoid"
+                ]
+            }
+        },
+        {
+            "name": "C",
+            "description": "Regularization parameter. The strength of the regularization is inversely proportional to C. Must be strictly positive.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "epsilon",
+            "description": "Epsilon in the epsilon-SVR model. Specifies the epsilon-tube within which no penalty is associated in the training loss function with points predicted within a distance epsilon from the actual value.",
+            "optional": true,
+            "default": 0.1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "gamma",
+            "description": "Kernel coefficient for 'rbf', 'poly', and 'sigmoid'. Higher values lead to tighter fits.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "degree",
+            "description": "Degree of the polynomial kernel function (only relevant for 'poly' kernel).",
+            "optional": true,
+            "default": 3,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "coef0",
+            "description": "Independent term in the kernel function (only relevant for 'poly' and 'sigmoid' kernels).",
+            "optional": true,
+            "default": 0,
+            "schema": {
+                "type": "number"
+            }
+        },
+        {
+            "name": "tolerance",
+            "description": "Tolerance of termination criterion.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "cachesize",
+            "description": "Size of the kernel cache in MB.",
+            "optional": true,
+            "default": 1000,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://link.springer.com/chapter/10.1007/978-1-4302-5990-9_4",
+            "title": "Awad, M., Khanna, R., Awad, M., & Khanna, R. (2015). Support vector regression. Efficient learning machines: Theories, concepts, and applications for engineers and system designers.",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals/mlm_class_catboost.json b/proposals/mlm_class_catboost.json
new file mode 100644
index 00000000..7a46d0a1
--- /dev/null
+++ b/proposals/mlm_class_catboost.json
@@ -0,0 +1,62 @@
+{
+    "id": "mlm_class_catboost",
+    "summary": "Initialize a CatBoost classification model",
+    "description": "Initializes a CatBoost classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "iterations",
+            "description": "The maximum number of trees that can be built during the training process.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 1,
+                "maximum": 500
+            }
+        },
+        {
+            "name": "depth",
+            "description": "Depth of the trees in the CatBoost model.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 1,
+                "maximum": 16
+            }
+        },
+        {
+            "name": "seed",
+            "description": "The random seed used for training, for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": 0,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ],
+                "minimum": 0,
+                "maximum": 2147483647
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://catboost.ai/",
+            "title": "CatBoost Documentation",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals/mlm_regr_svm.json b/proposals/mlm_regr_svm.json
new file mode 100644
index 00000000..ba63930c
--- /dev/null
+++ b/proposals/mlm_regr_svm.json
@@ -0,0 +1,122 @@
+{
+    "id": "mlm_regr_svm",
+    "summary": "Initialize an SVM regression model",
+    "description": "Initializes a Support Vector Machine (SVM) regression model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "kernel",
+            "description": "Specifies the kernel type to be used in the algorithm.",
+            "optional": true,
+            "default": "rbf",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "linear",
+                    "poly",
+                    "rbf",
+                    "sigmoid"
+                ]
+            }
+        },
+        {
+            "name": "C",
+            "description": "Regularization parameter. The strength of the regularization is inversely proportional to C. Must be strictly positive.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "epsilon",
+            "description": "Epsilon in the epsilon-SVR model. Specifies the epsilon-tube within which no penalty is associated in the training loss function with points predicted within a distance epsilon from the actual value.",
+            "optional": true,
+            "default": 0.1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "gamma",
+            "description": "Kernel coefficient for 'rbf', 'poly', and 'sigmoid'. Higher values lead to tighter fits.",
+            "optional": true,
+            "default": 1,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "degree",
+            "description": "Degree of the polynomial kernel function (only relevant for 'poly' kernel).",
+            "optional": true,
+            "default": 3,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "coef0",
+            "description": "Independent term in the kernel function (only relevant for 'poly' and 'sigmoid' kernels).",
+            "optional": true,
+            "default": 0,
+            "schema": {
+                "type": "number"
+            }
+        },
+        {
+            "name": "tolerance",
+            "description": "Tolerance of termination criterion.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "cachesize",
+            "description": "Size of the kernel cache in MB.",
+            "optional": true,
+            "default": 1000,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://link.springer.com/chapter/10.1007/978-1-4302-5990-9_4",
+            "title": "Awad, M., Khanna, R., Awad, M., & Khanna, R. (2015). Support vector regression. Efficient learning machines: Theories, concepts, and applications for engineers and system designers.",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/tests/.words b/tests/.words
index fb40804c..98ce6635 100644
--- a/tests/.words
+++ b/tests/.words
@@ -79,3 +79,7 @@ TAE
 least-confidence
 Camara
 softmax
+Khanna
+Awad
+CatBoost
+epsilon-SVR
\ No newline at end of file

From 13367b1c2ef41c481e46b823715165f9b38604ce Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Wed, 4 Jun 2025 13:58:57 +0200
Subject: [PATCH 31/52] update save ml and ml fit

---
 proposals-ml/ml_fit.json        |  2 +-
 proposals-ml/save_ml_model.json | 32 ++++++++++++++++++++++++++++++++
 proposals/ml_fit.json           |  2 +-
 proposals/save_ml_model.json    | 32 ++++++++++++++++++++++++++++++++
 4 files changed, 66 insertions(+), 2 deletions(-)

diff --git a/proposals-ml/ml_fit.json b/proposals-ml/ml_fit.json
index 2e01ff04..3825773c 100644
--- a/proposals-ml/ml_fit.json
+++ b/proposals-ml/ml_fit.json
@@ -47,7 +47,7 @@
         },
         {
             "name": "target",
-            "description": "The column in the training set that represents the dependent variable for model training.",
+            "description": "The name of the variable in the training set that serves as the target for model training.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals-ml/save_ml_model.json b/proposals-ml/save_ml_model.json
index a9622610..21787eda 100644
--- a/proposals-ml/save_ml_model.json
+++ b/proposals-ml/save_ml_model.json
@@ -16,6 +16,38 @@
                 "subtype": "ml-model"
             }
         },
+        {
+            "name": "name",
+            "description": "A distinct name of the model.",
+            "schema": {
+                "type": "string"
+            },
+            "required": true
+        },
+        {
+            "name": "tasks",
+            "description": "Specifies the ML tasks which the model can be used for.",
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "string",
+                    "enum": [
+                        "regression",
+                        "classification",
+                        "object-detection",
+                        "detection",
+                        "scene-classification",
+                        "segmentation",
+                        "semantic-segmentation",
+                        "similarity-search",
+                        "generative",
+                        "image-captioning",
+                        "super-resolution"
+                    ]
+                }
+            },
+            "required": true
+        },
         {
             "name": "options",
             "description": "Additional parameters to create the file(s).",
diff --git a/proposals/ml_fit.json b/proposals/ml_fit.json
index 2e01ff04..3825773c 100644
--- a/proposals/ml_fit.json
+++ b/proposals/ml_fit.json
@@ -47,7 +47,7 @@
         },
         {
             "name": "target",
-            "description": "The column in the training set that represents the dependent variable for model training.",
+            "description": "The name of the variable in the training set that serves as the target for model training.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals/save_ml_model.json b/proposals/save_ml_model.json
index a9622610..21787eda 100644
--- a/proposals/save_ml_model.json
+++ b/proposals/save_ml_model.json
@@ -16,6 +16,38 @@
                 "subtype": "ml-model"
             }
         },
+        {
+            "name": "name",
+            "description": "A distinct name of the model.",
+            "schema": {
+                "type": "string"
+            },
+            "required": true
+        },
+        {
+            "name": "tasks",
+            "description": "Specifies the ML tasks which the model can be used for.",
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "string",
+                    "enum": [
+                        "regression",
+                        "classification",
+                        "object-detection",
+                        "detection",
+                        "scene-classification",
+                        "segmentation",
+                        "semantic-segmentation",
+                        "similarity-search",
+                        "generative",
+                        "image-captioning",
+                        "super-resolution"
+                    ]
+                }
+            },
+            "required": true
+        },
         {
             "name": "options",
             "description": "Additional parameters to create the file(s).",

From 39dc49f9c127a4afd1d45a34028435e405a70bbb Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Wed, 4 Jun 2025 16:21:49 +0200
Subject: [PATCH 32/52] update save ml

---
 proposals-ml/save_ml_model.json | 6 ++----
 proposals/save_ml_model.json    | 6 ++----
 2 files changed, 4 insertions(+), 8 deletions(-)

diff --git a/proposals-ml/save_ml_model.json b/proposals-ml/save_ml_model.json
index 21787eda..af593363 100644
--- a/proposals-ml/save_ml_model.json
+++ b/proposals-ml/save_ml_model.json
@@ -21,8 +21,7 @@
             "description": "A distinct name of the model.",
             "schema": {
                 "type": "string"
-            },
-            "required": true
+            }
         },
         {
             "name": "tasks",
@@ -45,8 +44,7 @@
                         "super-resolution"
                     ]
                 }
-            },
-            "required": true
+            }
         },
         {
             "name": "options",
diff --git a/proposals/save_ml_model.json b/proposals/save_ml_model.json
index 21787eda..af593363 100644
--- a/proposals/save_ml_model.json
+++ b/proposals/save_ml_model.json
@@ -21,8 +21,7 @@
             "description": "A distinct name of the model.",
             "schema": {
                 "type": "string"
-            },
-            "required": true
+            }
         },
         {
             "name": "tasks",
@@ -45,8 +44,7 @@
                         "super-resolution"
                     ]
                 }
-            },
-            "required": true
+            }
         },
         {
             "name": "options",

From 1495b2c454c7162aaf30a1a4858579b114f9d840 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Thu, 5 Jun 2025 10:43:44 +0200
Subject: [PATCH 33/52] Dev (#11)

* ml classification smoothing

* smooth refactor

* smooth shorten summary

* ml label class

* softmax word

* 1d cnn

* svr and catboost class

* update save ml and ml fit

* update save ml
---
 proposals-ml/ml_fit.json        |  2 +-
 proposals-ml/save_ml_model.json | 30 ++++++++++++++++++++++++++++++
 proposals/ml_fit.json           |  2 +-
 proposals/save_ml_model.json    | 30 ++++++++++++++++++++++++++++++
 4 files changed, 62 insertions(+), 2 deletions(-)

diff --git a/proposals-ml/ml_fit.json b/proposals-ml/ml_fit.json
index 2e01ff04..3825773c 100644
--- a/proposals-ml/ml_fit.json
+++ b/proposals-ml/ml_fit.json
@@ -47,7 +47,7 @@
         },
         {
             "name": "target",
-            "description": "The column in the training set that represents the dependent variable for model training.",
+            "description": "The name of the variable in the training set that serves as the target for model training.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals-ml/save_ml_model.json b/proposals-ml/save_ml_model.json
index a9622610..af593363 100644
--- a/proposals-ml/save_ml_model.json
+++ b/proposals-ml/save_ml_model.json
@@ -16,6 +16,36 @@
                 "subtype": "ml-model"
             }
         },
+        {
+            "name": "name",
+            "description": "A distinct name of the model.",
+            "schema": {
+                "type": "string"
+            }
+        },
+        {
+            "name": "tasks",
+            "description": "Specifies the ML tasks which the model can be used for.",
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "string",
+                    "enum": [
+                        "regression",
+                        "classification",
+                        "object-detection",
+                        "detection",
+                        "scene-classification",
+                        "segmentation",
+                        "semantic-segmentation",
+                        "similarity-search",
+                        "generative",
+                        "image-captioning",
+                        "super-resolution"
+                    ]
+                }
+            }
+        },
         {
             "name": "options",
             "description": "Additional parameters to create the file(s).",
diff --git a/proposals/ml_fit.json b/proposals/ml_fit.json
index 2e01ff04..3825773c 100644
--- a/proposals/ml_fit.json
+++ b/proposals/ml_fit.json
@@ -47,7 +47,7 @@
         },
         {
             "name": "target",
-            "description": "The column in the training set that represents the dependent variable for model training.",
+            "description": "The name of the variable in the training set that serves as the target for model training.",
             "schema": {
                 "type": "string"
             }
diff --git a/proposals/save_ml_model.json b/proposals/save_ml_model.json
index a9622610..af593363 100644
--- a/proposals/save_ml_model.json
+++ b/proposals/save_ml_model.json
@@ -16,6 +16,36 @@
                 "subtype": "ml-model"
             }
         },
+        {
+            "name": "name",
+            "description": "A distinct name of the model.",
+            "schema": {
+                "type": "string"
+            }
+        },
+        {
+            "name": "tasks",
+            "description": "Specifies the ML tasks which the model can be used for.",
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "string",
+                    "enum": [
+                        "regression",
+                        "classification",
+                        "object-detection",
+                        "detection",
+                        "scene-classification",
+                        "segmentation",
+                        "semantic-segmentation",
+                        "similarity-search",
+                        "generative",
+                        "image-captioning",
+                        "super-resolution"
+                    ]
+                }
+            }
+        },
         {
             "name": "options",
             "description": "Additional parameters to create the file(s).",

From d83f94b30c91d5fc144046c24f3169353a22bdaa Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Thu, 3 Jul 2025 16:45:11 +0200
Subject: [PATCH 34/52] fit data mod

---
 proposals-ml/mlm_class_1dcnn.json | 178 ------------------------------
 proposals/ml_fit.json             |   3 +
 2 files changed, 3 insertions(+), 178 deletions(-)
 delete mode 100644 proposals-ml/mlm_class_1dcnn.json

diff --git a/proposals-ml/mlm_class_1dcnn.json b/proposals-ml/mlm_class_1dcnn.json
deleted file mode 100644
index 887276c0..00000000
--- a/proposals-ml/mlm_class_1dcnn.json
+++ /dev/null
@@ -1,178 +0,0 @@
-{
-    "id": "mlm_class_1dcnn",
-    "summary": "Initialize a 1D CNN classification model",
-    "description": "Initializes a 1D Convolutional Neural Network (CNN) classification model. The number of input channels and output classes are inferred automatically from the training data at fit time. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "conv_filters",
-            "description": "List of integers specifying the number of filters in each convolutional layer. The final output layer for classification will be added automatically based on the number of classes in the training data.",
-            "default": [64, 128, 256, 512],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "conv_kernels",
-            "description": "List of integers or 'global' for kernel size in each convolutional layer. The 4th conv uses a global kernel (covers the entire sequence at that layer). The final output layer for classification will use a 1x1 kernel and be added automatically.",
-            "default": [3, 3, 3, "global"],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "oneOf": [
-                        {"type": "integer", "minimum": 1},
-                        {"type": "string", "enum": ["global"]}
-                    ]
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "conv_strides",
-            "description": "List of integers specifying the stride for each convolutional layer. The final output layer for classification will use stride 1 and be added automatically.",
-            "default": [1, 1, 1, 1],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "use_batchnorm",
-            "description": "List of booleans specifying whether to use batch normalization after each convolutional layer. The final output layer will not use batch normalization.",
-            "default": [true, true, true, false],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "boolean"
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "activation",
-            "description": "Activation function to use after each convolutional layer.",
-            "default": "relu",
-            "schema": {
-                "type": "string",
-                "enum": ["relu", "tanh", "sigmoid", "leakyrelu"]
-            }
-        },
-        {
-            "name": "maxpool_sizes",
-            "description": "List of integers specifying the pool size for each max pooling layer (after each of the first 3 conv blocks).",
-            "default": [2, 2, 2],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 3,
-                "maxItems": 3
-            }
-        },
-        {
-            "name": "maxpool_strides",
-            "description": "List of integers specifying the stride for each max pooling layer.",
-            "default": [2, 2, 2],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 3,
-                "maxItems": 3
-            }
-        },
-        {
-            "name": "epochs",
-            "description": "Number of training epochs.",
-            "optional": true,
-            "default": 100,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "batch_size",
-            "description": "Size of the training batches.",
-            "optional": true,
-            "default": 64,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "optimizer",
-            "description": "The optimizer to use for training.",
-            "optional": true,
-            "default": "adam",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "adam",
-                    "sgd",
-                    "rmsprop",
-                    "adagrad",
-                    "nadam"
-                ]
-            }
-        },
-        {
-            "name": "learning_rate",
-            "description": "The learning rate for the optimizer.",
-            "optional": true,
-            "default": 0.001,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://ieeexplore.ieee.org/document/8921180",
-            "title": "Song et al. (2019): Land Cover Classification for Satellite Images Through 1D CNN",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-} 
\ No newline at end of file
diff --git a/proposals/ml_fit.json b/proposals/ml_fit.json
index 3825773c..e792bd64 100644
--- a/proposals/ml_fit.json
+++ b/proposals/ml_fit.json
@@ -42,6 +42,9 @@
                             "type": "other"
                         }
                     ]
+                },
+                {
+                    "type": "string"
                 }
             ]
         },

From 0cd3c493961788f12b7311a7c55e32d9c60d909f Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Mon, 7 Jul 2025 15:15:46 +0200
Subject: [PATCH 35/52] Dev (#12)

* ml classification smoothing

* smooth refactor

* smooth shorten summary

* ml label class

* softmax word

* 1d cnn

* svr and catboost class

* update save ml and ml fit

* update save ml

* fit data mod
---
 proposals/ml_fit.json | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/proposals/ml_fit.json b/proposals/ml_fit.json
index 3825773c..e792bd64 100644
--- a/proposals/ml_fit.json
+++ b/proposals/ml_fit.json
@@ -42,6 +42,9 @@
                             "type": "other"
                         }
                     ]
+                },
+                {
+                    "type": "string"
                 }
             ]
         },

From 20b3995a411006bc1277ea5635d6dd36b102c5b6 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Wed, 23 Jul 2025 14:49:38 +0200
Subject: [PATCH 36/52] Dev (#13)

* ml classification smoothing

* smooth refactor

* smooth shorten summary

* ml label class

* softmax word

* 1d cnn

* svr and catboost class

* update save ml and ml fit

* update save ml

* fit data mod

From 5d43c9ed63036186341b820746369040fc1269a0 Mon Sep 17 00:00:00 2001
From: = <jonhur@web.de>
Date: Tue, 16 Sep 2025 14:07:53 +0200
Subject: [PATCH 37/52] added input_index parameter

---
 proposals-ml/load_ml_model.json | 9 +++++++++
 proposals/load_ml_model.json    | 9 +++++++++
 2 files changed, 18 insertions(+)

diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_ml_model.json
index bbc69d4b..956dd473 100644
--- a/proposals-ml/load_ml_model.json
+++ b/proposals-ml/load_ml_model.json
@@ -35,6 +35,15 @@
             },
             "default": null,
             "optinoal": true
+        },
+        {
+            "name": "input_index",
+            "description": "STAC:MLM supports multiple ML model input specification. This parameter specifies the index of the input specification in the ``mlm:input`` array to use for prediction or training. As this is an array, the first input in the array has index 0.",
+            "schema": {
+                "type": "integer"
+            },
+            "default": 0,
+            "optional": true
         }
     ],
     "returns": {
diff --git a/proposals/load_ml_model.json b/proposals/load_ml_model.json
index bbc69d4b..956dd473 100644
--- a/proposals/load_ml_model.json
+++ b/proposals/load_ml_model.json
@@ -35,6 +35,15 @@
             },
             "default": null,
             "optinoal": true
+        },
+        {
+            "name": "input_index",
+            "description": "STAC:MLM supports multiple ML model input specification. This parameter specifies the index of the input specification in the ``mlm:input`` array to use for prediction or training. As this is an array, the first input in the array has index 0.",
+            "schema": {
+                "type": "integer"
+            },
+            "default": 0,
+            "optional": true
         }
     ],
     "returns": {

From 184b6751cf986ba63e233b1b5e0bbc9fd8a25914 Mon Sep 17 00:00:00 2001
From: = <jonhur@web.de>
Date: Tue, 16 Sep 2025 14:09:05 +0200
Subject: [PATCH 38/52] fixed typo

---
 proposals-ml/load_ml_model.json | 2 +-
 proposals/load_ml_model.json    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_ml_model.json
index 956dd473..07def956 100644
--- a/proposals-ml/load_ml_model.json
+++ b/proposals-ml/load_ml_model.json
@@ -34,7 +34,7 @@
                 "type": "string"
             },
             "default": null,
-            "optinoal": true
+            "optional": true
         },
         {
             "name": "input_index",
diff --git a/proposals/load_ml_model.json b/proposals/load_ml_model.json
index 956dd473..07def956 100644
--- a/proposals/load_ml_model.json
+++ b/proposals/load_ml_model.json
@@ -34,7 +34,7 @@
                 "type": "string"
             },
             "default": null,
-            "optinoal": true
+            "optional": true
         },
         {
             "name": "input_index",

From 0f8ff1f1fd52754f3a93726996c2e7aae2ecb48d Mon Sep 17 00:00:00 2001
From: = <jonhur@web.de>
Date: Tue, 16 Sep 2025 14:34:16 +0200
Subject: [PATCH 39/52] added output_index param

---
 proposals-ml/load_ml_model.json | 9 +++++++++
 proposals/load_ml_model.json    | 9 +++++++++
 2 files changed, 18 insertions(+)

diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_ml_model.json
index 07def956..489a0ef9 100644
--- a/proposals-ml/load_ml_model.json
+++ b/proposals-ml/load_ml_model.json
@@ -44,6 +44,15 @@
             },
             "default": 0,
             "optional": true
+        },
+        {
+            "name": "output_index",
+            "description": "STAC:MLM items supports multiple ML model output specification. This parameter specifies the index of the output specification in the ``mlm:output`` array to use for prediction or training. As ``mlm:output`` is an array, the first output in the array has index 0.",
+            "schema": {
+                "type": "integer"
+            },
+            "default": 0,
+            "optional": true
         }
     ],
     "returns": {
diff --git a/proposals/load_ml_model.json b/proposals/load_ml_model.json
index 07def956..489a0ef9 100644
--- a/proposals/load_ml_model.json
+++ b/proposals/load_ml_model.json
@@ -44,6 +44,15 @@
             },
             "default": 0,
             "optional": true
+        },
+        {
+            "name": "output_index",
+            "description": "STAC:MLM items supports multiple ML model output specification. This parameter specifies the index of the output specification in the ``mlm:output`` array to use for prediction or training. As ``mlm:output`` is an array, the first output in the array has index 0.",
+            "schema": {
+                "type": "integer"
+            },
+            "default": 0,
+            "optional": true
         }
     ],
     "returns": {

From 80c0c20f33f7c523799c047cbb6fc65ac7ca5492 Mon Sep 17 00:00:00 2001
From: = <jonhur@web.de>
Date: Tue, 16 Sep 2025 14:36:24 +0200
Subject: [PATCH 40/52] refined description

---
 proposals-ml/load_ml_model.json | 2 +-
 proposals/load_ml_model.json    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_ml_model.json
index 489a0ef9..6c674b43 100644
--- a/proposals-ml/load_ml_model.json
+++ b/proposals-ml/load_ml_model.json
@@ -38,7 +38,7 @@
         },
         {
             "name": "input_index",
-            "description": "STAC:MLM supports multiple ML model input specification. This parameter specifies the index of the input specification in the ``mlm:input`` array to use for prediction or training. As this is an array, the first input in the array has index 0.",
+            "description": "STAC:MLM items supports multiple ML model input specification. This parameter specifies the index of the input specification in the ``mlm:input`` array to use for prediction or training. As ``mlm:input`` is an array, the first input in the array has index 0.",
             "schema": {
                 "type": "integer"
             },
diff --git a/proposals/load_ml_model.json b/proposals/load_ml_model.json
index 489a0ef9..6c674b43 100644
--- a/proposals/load_ml_model.json
+++ b/proposals/load_ml_model.json
@@ -38,7 +38,7 @@
         },
         {
             "name": "input_index",
-            "description": "STAC:MLM supports multiple ML model input specification. This parameter specifies the index of the input specification in the ``mlm:input`` array to use for prediction or training. As this is an array, the first input in the array has index 0.",
+            "description": "STAC:MLM items supports multiple ML model input specification. This parameter specifies the index of the input specification in the ``mlm:input`` array to use for prediction or training. As ``mlm:input`` is an array, the first input in the array has index 0.",
             "schema": {
                 "type": "integer"
             },

From 8b9b70f120f1022d04a8d751c4c60da1b4910fcb Mon Sep 17 00:00:00 2001
From: Jonas Hurst <76062450+jonas-hurst@users.noreply.github.com>
Date: Wed, 24 Sep 2025 15:08:46 +0200
Subject: [PATCH 41/52] removed dimensions parameter (#18)

* svr and catboost (#10)

* ml classification smoothing

* smooth refactor

* smooth shorten summary

* ml label class

* softmax word

* 1d cnn

* svr and catboost class

* Dev (#11)

* ml classification smoothing

* smooth refactor

* smooth shorten summary

* ml label class

* softmax word

* 1d cnn

* svr and catboost class

* update save ml and ml fit

* update save ml

* Dev (#12)

* ml classification smoothing

* smooth refactor

* smooth shorten summary

* ml label class

* softmax word

* 1d cnn

* svr and catboost class

* update save ml and ml fit

* update save ml

* fit data mod

* Dev (#13)

* ml classification smoothing

* smooth refactor

* smooth shorten summary

* ml label class

* softmax word

* 1d cnn

* svr and catboost class

* update save ml and ml fit

* update save ml

* fit data mod

* removed dimensions parameter

---------

Co-authored-by: Brian Pondi <pondibrian@gmail.com>
---
 proposals-ml/ml_predict.json               |  10 --
 proposals-ml/ml_predict_probabilities.json |  10 --
 proposals-ml/mlm_class_1dcnn.json          | 178 +++++++++++++++++++++
 proposals/ml_predict.json                  |  10 --
 proposals/ml_predict_probabilities.json    |  10 --
 5 files changed, 178 insertions(+), 40 deletions(-)
 create mode 100644 proposals-ml/mlm_class_1dcnn.json

diff --git a/proposals-ml/ml_predict.json b/proposals-ml/ml_predict.json
index 3954e95b..4741bcd9 100644
--- a/proposals-ml/ml_predict.json
+++ b/proposals-ml/ml_predict.json
@@ -22,16 +22,6 @@
                 "type": "object",
                 "subtype": "ml-model"
             }
-        },
-        {
-            "name": "dimensions",
-            "description": "Zero or more dimensions that will be reduced by the model. Fails with a `DimensionNotAvailable` exception if one of the specified dimensions does not exist.",
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "string"
-                }
-            }
         }
     ],
     "returns": {
diff --git a/proposals-ml/ml_predict_probabilities.json b/proposals-ml/ml_predict_probabilities.json
index 68cc78c7..c44689b6 100644
--- a/proposals-ml/ml_predict_probabilities.json
+++ b/proposals-ml/ml_predict_probabilities.json
@@ -22,16 +22,6 @@
                 "type": "object",
                 "subtype": "ml-model"
             }
-        },
-        {
-            "name": "dimensions",
-            "description": "Zero or more dimensions that will be reduced by the model. Fails with a `DimensionNotAvailable` exception if one of the specified dimensions does not exist.",
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "string"
-                }
-            }
         }
     ],
     "returns": {
diff --git a/proposals-ml/mlm_class_1dcnn.json b/proposals-ml/mlm_class_1dcnn.json
new file mode 100644
index 00000000..887276c0
--- /dev/null
+++ b/proposals-ml/mlm_class_1dcnn.json
@@ -0,0 +1,178 @@
+{
+    "id": "mlm_class_1dcnn",
+    "summary": "Initialize a 1D CNN classification model",
+    "description": "Initializes a 1D Convolutional Neural Network (CNN) classification model. The number of input channels and output classes are inferred automatically from the training data at fit time. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "conv_filters",
+            "description": "List of integers specifying the number of filters in each convolutional layer. The final output layer for classification will be added automatically based on the number of classes in the training data.",
+            "default": [64, 128, 256, 512],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "conv_kernels",
+            "description": "List of integers or 'global' for kernel size in each convolutional layer. The 4th conv uses a global kernel (covers the entire sequence at that layer). The final output layer for classification will use a 1x1 kernel and be added automatically.",
+            "default": [3, 3, 3, "global"],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "oneOf": [
+                        {"type": "integer", "minimum": 1},
+                        {"type": "string", "enum": ["global"]}
+                    ]
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "conv_strides",
+            "description": "List of integers specifying the stride for each convolutional layer. The final output layer for classification will use stride 1 and be added automatically.",
+            "default": [1, 1, 1, 1],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "use_batchnorm",
+            "description": "List of booleans specifying whether to use batch normalization after each convolutional layer. The final output layer will not use batch normalization.",
+            "default": [true, true, true, false],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "boolean"
+                },
+                "minItems": 4,
+                "maxItems": 4
+            }
+        },
+        {
+            "name": "activation",
+            "description": "Activation function to use after each convolutional layer.",
+            "default": "relu",
+            "schema": {
+                "type": "string",
+                "enum": ["relu", "tanh", "sigmoid", "leakyrelu"]
+            }
+        },
+        {
+            "name": "maxpool_sizes",
+            "description": "List of integers specifying the pool size for each max pooling layer (after each of the first 3 conv blocks).",
+            "default": [2, 2, 2],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 3,
+                "maxItems": 3
+            }
+        },
+        {
+            "name": "maxpool_strides",
+            "description": "List of integers specifying the stride for each max pooling layer.",
+            "default": [2, 2, 2],
+            "schema": {
+                "type": "array",
+                "items": {
+                    "type": "integer",
+                    "minimum": 1
+                },
+                "minItems": 3,
+                "maxItems": 3
+            }
+        },
+        {
+            "name": "epochs",
+            "description": "Number of training epochs.",
+            "optional": true,
+            "default": 100,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "batch_size",
+            "description": "Size of the training batches.",
+            "optional": true,
+            "default": 64,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "optimizer",
+            "description": "The optimizer to use for training.",
+            "optional": true,
+            "default": "adam",
+            "schema": {
+                "type": "string",
+                "enum": [
+                    "adam",
+                    "sgd",
+                    "rmsprop",
+                    "adagrad",
+                    "nadam"
+                ]
+            }
+        },
+        {
+            "name": "learning_rate",
+            "description": "The learning rate for the optimizer.",
+            "optional": true,
+            "default": 0.001,
+            "schema": {
+                "type": "number",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "seed",
+            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A model object that can be trained using ``ml_fit``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://ieeexplore.ieee.org/document/8921180",
+            "title": "Song et al. (2019): Land Cover Classification for Satellite Images Through 1D CNN",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+} 
\ No newline at end of file
diff --git a/proposals/ml_predict.json b/proposals/ml_predict.json
index 3954e95b..4741bcd9 100644
--- a/proposals/ml_predict.json
+++ b/proposals/ml_predict.json
@@ -22,16 +22,6 @@
                 "type": "object",
                 "subtype": "ml-model"
             }
-        },
-        {
-            "name": "dimensions",
-            "description": "Zero or more dimensions that will be reduced by the model. Fails with a `DimensionNotAvailable` exception if one of the specified dimensions does not exist.",
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "string"
-                }
-            }
         }
     ],
     "returns": {
diff --git a/proposals/ml_predict_probabilities.json b/proposals/ml_predict_probabilities.json
index 68cc78c7..c44689b6 100644
--- a/proposals/ml_predict_probabilities.json
+++ b/proposals/ml_predict_probabilities.json
@@ -22,16 +22,6 @@
                 "type": "object",
                 "subtype": "ml-model"
             }
-        },
-        {
-            "name": "dimensions",
-            "description": "Zero or more dimensions that will be reduced by the model. Fails with a `DimensionNotAvailable` exception if one of the specified dimensions does not exist.",
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "string"
-                }
-            }
         }
     ],
     "returns": {

From 5505774dff0f09beab8ca46d07a7b64fd335216f Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Wed, 24 Sep 2025 15:19:41 +0200
Subject: [PATCH 42/52] 1d cnn removed

---
 proposals-ml/mlm_class_1dcnn.json | 178 ------------------------------
 1 file changed, 178 deletions(-)
 delete mode 100644 proposals-ml/mlm_class_1dcnn.json

diff --git a/proposals-ml/mlm_class_1dcnn.json b/proposals-ml/mlm_class_1dcnn.json
deleted file mode 100644
index 887276c0..00000000
--- a/proposals-ml/mlm_class_1dcnn.json
+++ /dev/null
@@ -1,178 +0,0 @@
-{
-    "id": "mlm_class_1dcnn",
-    "summary": "Initialize a 1D CNN classification model",
-    "description": "Initializes a 1D Convolutional Neural Network (CNN) classification model. The number of input channels and output classes are inferred automatically from the training data at fit time. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "conv_filters",
-            "description": "List of integers specifying the number of filters in each convolutional layer. The final output layer for classification will be added automatically based on the number of classes in the training data.",
-            "default": [64, 128, 256, 512],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "conv_kernels",
-            "description": "List of integers or 'global' for kernel size in each convolutional layer. The 4th conv uses a global kernel (covers the entire sequence at that layer). The final output layer for classification will use a 1x1 kernel and be added automatically.",
-            "default": [3, 3, 3, "global"],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "oneOf": [
-                        {"type": "integer", "minimum": 1},
-                        {"type": "string", "enum": ["global"]}
-                    ]
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "conv_strides",
-            "description": "List of integers specifying the stride for each convolutional layer. The final output layer for classification will use stride 1 and be added automatically.",
-            "default": [1, 1, 1, 1],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "use_batchnorm",
-            "description": "List of booleans specifying whether to use batch normalization after each convolutional layer. The final output layer will not use batch normalization.",
-            "default": [true, true, true, false],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "boolean"
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "activation",
-            "description": "Activation function to use after each convolutional layer.",
-            "default": "relu",
-            "schema": {
-                "type": "string",
-                "enum": ["relu", "tanh", "sigmoid", "leakyrelu"]
-            }
-        },
-        {
-            "name": "maxpool_sizes",
-            "description": "List of integers specifying the pool size for each max pooling layer (after each of the first 3 conv blocks).",
-            "default": [2, 2, 2],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 3,
-                "maxItems": 3
-            }
-        },
-        {
-            "name": "maxpool_strides",
-            "description": "List of integers specifying the stride for each max pooling layer.",
-            "default": [2, 2, 2],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 3,
-                "maxItems": 3
-            }
-        },
-        {
-            "name": "epochs",
-            "description": "Number of training epochs.",
-            "optional": true,
-            "default": 100,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "batch_size",
-            "description": "Size of the training batches.",
-            "optional": true,
-            "default": 64,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "optimizer",
-            "description": "The optimizer to use for training.",
-            "optional": true,
-            "default": "adam",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "adam",
-                    "sgd",
-                    "rmsprop",
-                    "adagrad",
-                    "nadam"
-                ]
-            }
-        },
-        {
-            "name": "learning_rate",
-            "description": "The learning rate for the optimizer.",
-            "optional": true,
-            "default": 0.001,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://ieeexplore.ieee.org/document/8921180",
-            "title": "Song et al. (2019): Land Cover Classification for Satellite Images Through 1D CNN",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-} 
\ No newline at end of file

From 29dbd6bab99bb95ea31c31dbb87d39123c4325ec Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Thu, 9 Oct 2025 10:36:06 +0200
Subject: [PATCH 43/52] load stac ml

---
 .../{load_ml_model.json => load_stac_ml.json} |   6 +-
 proposals-ml/mlm_class_1dcnn.json             | 178 ------------------
 2 files changed, 3 insertions(+), 181 deletions(-)
 rename proposals-ml/{load_ml_model.json => load_stac_ml.json} (96%)
 delete mode 100644 proposals-ml/mlm_class_1dcnn.json

diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_stac_ml.json
similarity index 96%
rename from proposals-ml/load_ml_model.json
rename to proposals-ml/load_stac_ml.json
index 6c674b43..f776bc06 100644
--- a/proposals-ml/load_ml_model.json
+++ b/proposals-ml/load_stac_ml.json
@@ -1,7 +1,7 @@
 {
-    "id": "load_ml_model",
-    "summary": "Load a ML model",
-    "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()``.",
+    "id": "load_stac_ml",
+    "summary": "Load a ML model from a STAC Item",
+    "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()`` or  externally hosted models.",
     "categories": [
         "machine learning",
         "import"
diff --git a/proposals-ml/mlm_class_1dcnn.json b/proposals-ml/mlm_class_1dcnn.json
deleted file mode 100644
index 887276c0..00000000
--- a/proposals-ml/mlm_class_1dcnn.json
+++ /dev/null
@@ -1,178 +0,0 @@
-{
-    "id": "mlm_class_1dcnn",
-    "summary": "Initialize a 1D CNN classification model",
-    "description": "Initializes a 1D Convolutional Neural Network (CNN) classification model. The number of input channels and output classes are inferred automatically from the training data at fit time. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "conv_filters",
-            "description": "List of integers specifying the number of filters in each convolutional layer. The final output layer for classification will be added automatically based on the number of classes in the training data.",
-            "default": [64, 128, 256, 512],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "conv_kernels",
-            "description": "List of integers or 'global' for kernel size in each convolutional layer. The 4th conv uses a global kernel (covers the entire sequence at that layer). The final output layer for classification will use a 1x1 kernel and be added automatically.",
-            "default": [3, 3, 3, "global"],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "oneOf": [
-                        {"type": "integer", "minimum": 1},
-                        {"type": "string", "enum": ["global"]}
-                    ]
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "conv_strides",
-            "description": "List of integers specifying the stride for each convolutional layer. The final output layer for classification will use stride 1 and be added automatically.",
-            "default": [1, 1, 1, 1],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "use_batchnorm",
-            "description": "List of booleans specifying whether to use batch normalization after each convolutional layer. The final output layer will not use batch normalization.",
-            "default": [true, true, true, false],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "boolean"
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "activation",
-            "description": "Activation function to use after each convolutional layer.",
-            "default": "relu",
-            "schema": {
-                "type": "string",
-                "enum": ["relu", "tanh", "sigmoid", "leakyrelu"]
-            }
-        },
-        {
-            "name": "maxpool_sizes",
-            "description": "List of integers specifying the pool size for each max pooling layer (after each of the first 3 conv blocks).",
-            "default": [2, 2, 2],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 3,
-                "maxItems": 3
-            }
-        },
-        {
-            "name": "maxpool_strides",
-            "description": "List of integers specifying the stride for each max pooling layer.",
-            "default": [2, 2, 2],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 3,
-                "maxItems": 3
-            }
-        },
-        {
-            "name": "epochs",
-            "description": "Number of training epochs.",
-            "optional": true,
-            "default": 100,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "batch_size",
-            "description": "Size of the training batches.",
-            "optional": true,
-            "default": 64,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "optimizer",
-            "description": "The optimizer to use for training.",
-            "optional": true,
-            "default": "adam",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "adam",
-                    "sgd",
-                    "rmsprop",
-                    "adagrad",
-                    "nadam"
-                ]
-            }
-        },
-        {
-            "name": "learning_rate",
-            "description": "The learning rate for the optimizer.",
-            "optional": true,
-            "default": 0.001,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://ieeexplore.ieee.org/document/8921180",
-            "title": "Song et al. (2019): Land Cover Classification for Satellite Images Through 1D CNN",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-} 
\ No newline at end of file

From 49f10099e2a39886f1063edded491061df2b4386 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Thu, 9 Oct 2025 10:43:47 +0200
Subject: [PATCH 44/52] remove unused export and import processes

---
 proposals-ml/export_ml_model.json             | 40 -------------------
 proposals-ml/import_ml_model.json             | 33 ---------------
 proposals/export_ml_model.json                | 40 -------------------
 proposals/import_ml_model.json                | 33 ---------------
 .../{load_ml_model.json => load_stac_ml.json} |  7 ++--
 5 files changed, 3 insertions(+), 150 deletions(-)
 delete mode 100644 proposals-ml/export_ml_model.json
 delete mode 100644 proposals-ml/import_ml_model.json
 delete mode 100644 proposals/export_ml_model.json
 delete mode 100644 proposals/import_ml_model.json
 rename proposals/{load_ml_model.json => load_stac_ml.json} (96%)

diff --git a/proposals-ml/export_ml_model.json b/proposals-ml/export_ml_model.json
deleted file mode 100644
index cbf6fffa..00000000
--- a/proposals-ml/export_ml_model.json
+++ /dev/null
@@ -1,40 +0,0 @@
-{
-    "id": "export_ml_model",
-    "summary": "Export a machine learning model to a workspace",
-    "description": "Exports a machine learning model to a specified workspace location. This is designed for intermediate model storage and can use the implementation's native serialization format.\n\nUnlike save_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for temporary storage and transfer of models between processes.",
-    "categories": [
-        "Export",
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "model",
-            "description": "The machine learning model to be exported. This can be in any format supported by the implementation.",
-            "schema": {
-                "type": "object",
-                "subtype": "ml-model"
-            }
-        },
-        {
-            "name": "name",
-            "description": "The name to be given to the exported model. This will be used to identify the model when importing it later.",
-            "schema": {
-                "type": "string"
-            }
-        },
-        {
-            "name": "folder",
-            "description": "The folder path within the workspace where the model should be stored.",
-            "schema": {
-                "type": "string"
-            }
-        }
-    ],
-    "returns": {
-        "description": "A boolean indicating whether the export was successful.",
-        "schema": {
-            "type": "boolean"
-        }
-    }
-} 
\ No newline at end of file
diff --git a/proposals-ml/import_ml_model.json b/proposals-ml/import_ml_model.json
deleted file mode 100644
index 5ba57f61..00000000
--- a/proposals-ml/import_ml_model.json
+++ /dev/null
@@ -1,33 +0,0 @@
-{
-    "id": "import_ml_model",
-    "summary": "Import a machine learning model from a workspace",
-    "description": "Imports a previously exported machine learning model from a specified workspace location. This is designed for intermediate model retrieval and can handle models in the implementation's native serialization format.\n\nUnlike load_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for retrieving temporarily stored models between processes.",
-    "categories": [
-        "Import",
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "name",
-            "description": "The name of the model to import. This should match the name used when the model was exported.",
-            "schema": {
-                "type": "string"
-            }
-        },
-        {
-            "name": "folder",
-            "description": "The folder path within the workspace where the model is stored.",
-            "schema": {
-                "type": "string"
-            }
-        }
-    ],
-    "returns": {
-        "description": "The imported machine learning model in its original format.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    }
-} 
\ No newline at end of file
diff --git a/proposals/export_ml_model.json b/proposals/export_ml_model.json
deleted file mode 100644
index cbf6fffa..00000000
--- a/proposals/export_ml_model.json
+++ /dev/null
@@ -1,40 +0,0 @@
-{
-    "id": "export_ml_model",
-    "summary": "Export a machine learning model to a workspace",
-    "description": "Exports a machine learning model to a specified workspace location. This is designed for intermediate model storage and can use the implementation's native serialization format.\n\nUnlike save_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for temporary storage and transfer of models between processes.",
-    "categories": [
-        "Export",
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "model",
-            "description": "The machine learning model to be exported. This can be in any format supported by the implementation.",
-            "schema": {
-                "type": "object",
-                "subtype": "ml-model"
-            }
-        },
-        {
-            "name": "name",
-            "description": "The name to be given to the exported model. This will be used to identify the model when importing it later.",
-            "schema": {
-                "type": "string"
-            }
-        },
-        {
-            "name": "folder",
-            "description": "The folder path within the workspace where the model should be stored.",
-            "schema": {
-                "type": "string"
-            }
-        }
-    ],
-    "returns": {
-        "description": "A boolean indicating whether the export was successful.",
-        "schema": {
-            "type": "boolean"
-        }
-    }
-} 
\ No newline at end of file
diff --git a/proposals/import_ml_model.json b/proposals/import_ml_model.json
deleted file mode 100644
index 5ba57f61..00000000
--- a/proposals/import_ml_model.json
+++ /dev/null
@@ -1,33 +0,0 @@
-{
-    "id": "import_ml_model",
-    "summary": "Import a machine learning model from a workspace",
-    "description": "Imports a previously exported machine learning model from a specified workspace location. This is designed for intermediate model retrieval and can handle models in the implementation's native serialization format.\n\nUnlike load_ml_model, this process doesn't require STAC compliance or specific model formats (like ONNX), making it suitable for retrieving temporarily stored models between processes.",
-    "categories": [
-        "Import",
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "name",
-            "description": "The name of the model to import. This should match the name used when the model was exported.",
-            "schema": {
-                "type": "string"
-            }
-        },
-        {
-            "name": "folder",
-            "description": "The folder path within the workspace where the model is stored.",
-            "schema": {
-                "type": "string"
-            }
-        }
-    ],
-    "returns": {
-        "description": "The imported machine learning model in its original format.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    }
-} 
\ No newline at end of file
diff --git a/proposals/load_ml_model.json b/proposals/load_stac_ml.json
similarity index 96%
rename from proposals/load_ml_model.json
rename to proposals/load_stac_ml.json
index 6c674b43..d3c6426f 100644
--- a/proposals/load_ml_model.json
+++ b/proposals/load_stac_ml.json
@@ -1,7 +1,7 @@
 {
-    "id": "load_ml_model",
-    "summary": "Load a ML model",
-    "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()``.",
+    "id": "load_stac_ml",
+    "summary": "Load a ML model from a STAC Item",
+    "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()`` or  externally hosted models.",
     "categories": [
         "machine learning",
         "import"
@@ -71,4 +71,3 @@
         }
     ]
 }
-

From bb17435c752844373a3056458d0574259ea6c541 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Thu, 9 Oct 2025 11:27:17 +0200
Subject: [PATCH 45/52] load ml model

---
 proposals-ml/load_ml_model.json | 43 +++++++++++++++++++++++++++++++++
 proposals/load_ml_model.json    | 43 +++++++++++++++++++++++++++++++++
 tests/.words                    |  3 ++-
 3 files changed, 88 insertions(+), 1 deletion(-)
 create mode 100644 proposals-ml/load_ml_model.json
 create mode 100644 proposals/load_ml_model.json

diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_ml_model.json
new file mode 100644
index 00000000..8980a20d
--- /dev/null
+++ b/proposals-ml/load_ml_model.json
@@ -0,0 +1,43 @@
+{
+    "id": "load_ml_model",
+    "summary": "Load a machine learning model by ID",
+    "description": "Loads a machine learning model that is managed by the current back-end, identified by a back-end specific model identifier.\n\nThis allows the back-end to host and optimize models for efficient inference and training (e.g., via model-specific libraries such as Terratorch). Back-end specific models can be referenced by a string identifier similar to collections.\n\nIf you want to load a model from a STAC Item implementing the `mlm` extension, use `load_stac_ml()` instead.",
+    "categories": [
+        "machine learning",
+        "import"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "id",
+            "description": "The back-end specific identifier of the machine learning model to load.",
+            "schema": {
+                "type": "string",
+                "pattern": "^[\\w\\-\\.~/]+$"
+            }
+        }
+    ],
+    "returns": {
+        "description": "A machine learning model to be used with machine learning processes such as `ml_predict()`.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://github.com/stac-extensions/mlm",
+            "title": "Machine Learning Model STAC extension",
+            "type": "text/html",
+            "rel": "about"
+        },
+        {
+            "href": "https://openeo.org/documentation/1.0/",
+            "title": "openEO API documentation",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
+
+
diff --git a/proposals/load_ml_model.json b/proposals/load_ml_model.json
new file mode 100644
index 00000000..8980a20d
--- /dev/null
+++ b/proposals/load_ml_model.json
@@ -0,0 +1,43 @@
+{
+    "id": "load_ml_model",
+    "summary": "Load a machine learning model by ID",
+    "description": "Loads a machine learning model that is managed by the current back-end, identified by a back-end specific model identifier.\n\nThis allows the back-end to host and optimize models for efficient inference and training (e.g., via model-specific libraries such as Terratorch). Back-end specific models can be referenced by a string identifier similar to collections.\n\nIf you want to load a model from a STAC Item implementing the `mlm` extension, use `load_stac_ml()` instead.",
+    "categories": [
+        "machine learning",
+        "import"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "id",
+            "description": "The back-end specific identifier of the machine learning model to load.",
+            "schema": {
+                "type": "string",
+                "pattern": "^[\\w\\-\\.~/]+$"
+            }
+        }
+    ],
+    "returns": {
+        "description": "A machine learning model to be used with machine learning processes such as `ml_predict()`.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://github.com/stac-extensions/mlm",
+            "title": "Machine Learning Model STAC extension",
+            "type": "text/html",
+            "rel": "about"
+        },
+        {
+            "href": "https://openeo.org/documentation/1.0/",
+            "title": "openEO API documentation",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
+
+
diff --git a/tests/.words b/tests/.words
index 98ce6635..25ba9e5e 100644
--- a/tests/.words
+++ b/tests/.words
@@ -82,4 +82,5 @@ softmax
 Khanna
 Awad
 CatBoost
-epsilon-SVR
\ No newline at end of file
+epsilon-SVR
+Terratorch
\ No newline at end of file

From a7263c616eecccc3433fc77eb60cd237cd8e4299 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Tue, 21 Oct 2025 15:05:10 +0200
Subject: [PATCH 46/52] change log update

---
 CHANGELOG.md              |   3 +-
 proposals-ml/ML-README.md | 159 --------------------------------------
 2 files changed, 1 insertion(+), 161 deletions(-)
 delete mode 100644 proposals-ml/ML-README.md

diff --git a/CHANGELOG.md b/CHANGELOG.md
index b07d8349..7402db23 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -13,14 +13,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - New processes in proposal state:
     - `date_between`
     - `date_difference`
-    - `export_cube`
-    - `export_ml_model`
     - `filter_vector`
     - `flatten_dimensions`
     - `import_cube`
     - `import_ml_model`
     - `load_geojson`
     - `load_ml_model`
+    - `load_stac_ml`
     - `load_url`
     - `mlm_class_catboost`
     - `mlm_class_lighttae`
diff --git a/proposals-ml/ML-README.md b/proposals-ml/ML-README.md
deleted file mode 100644
index 8fad28fc..00000000
--- a/proposals-ml/ML-README.md
+++ /dev/null
@@ -1,159 +0,0 @@
-# ML API Specification for Earth Observation Data Cubes
-
-Welcome to the Machine Learning (ML) API for processing Earth Observation (EO) data cubes. This API provides capabilities to perform classification, regression, smoothing, uncertainty analysis, and more on EO data. Below is a comprehensive guide to the API's operations.
-
-
-
-## API Operations
-
-Here's a detailed breakdown of the ML processes available in this API.
-
-### General ML Processes
-
-- `mlm_class_*` and `mlm_regr_*`
-  - **Purpose**: Placeholder processes for classification and regression model definition.
-  - **Usage**: These processes encapsulate various machine learning model .
-
-- `ml_fit(model, training_set, target = "label")`
-  - **Purpose**: Train an ML model using the specified training dataset.
-  - **Parameters**:
-    - `model`: The machine learning algorithm/model to be trained.
-    - `training_set`: The dataset used for training.
-    - `target`: The label or target variable in the dataset.
-
-- `ml_predict(data, model)`
-  - **Purpose**: Predict outcomes using a trained model on a given dataset.
-  - **Parameters**:
-    - `data`: The dataset for prediction.
-    - `model`: The trained model to use for prediction.
-
-- `ml_predict_probability(data, model)`
-  - **Purpose**: Similar to `ml_predict`, but returns the probability of each prediction.
-  - **Parameters**:
-    - `data`: The dataset for probabilistic prediction.
-    - `model`: The trained model to use for prediction.
-
-- `ml_smooth_class(data, window_size, neighborhood_fraction)`
-  - **Purpose**: Smooth classification results with spatial filtering.
-  - **Parameters**:
-    - `data`: The dataset to be smoothed.
-    - `window_size`: The size of the moving window.
-    - `neighborhood_fraction`: The fraction of neighbors to consider.
-
-- `ml_uncertainty_class(data, method = "margin")`
-  - **Purpose**: Analyze and assess uncertainty in classification results.
-  - **Parameters**:
-    - `data`: The dataset for uncertainty assessment.
-    - `method`: The method for uncertainty evaluation. Default is "margin".
-
-### Model Management
-
-- `save_ml_model(data, name, tasks = list(), options = {})`
-  - **Purpose**: Save a trained model to persistent storage.
-  - **Parameters**:
-    - `data`: The model data to be saved.
-    - `name`: The identifier for the model.
-    - `tasks`: (Optional) A list of tasks associated with the model.
-    - `options`: (Optional) Additional options for saving the model.
-
-- `load_ml_model(uri)`
-  - **Purpose**: Load a previously saved model from storage.
-  - **Parameters**:
-    - `uri`: The location of the stored model.
-
-- `export_model(model, name, folder)`
-  - **Purpose**: Export a model to a specified folder in user workspace.
-  - **Parameters**:
-    - `model`: The model to export.
-    - `name`: The name to save the model as.
-    - `folder`: The destination folder.
-
-- `import_model(name, folder)`
-  - **Purpose**: Import a model from a specified folder in user workspace.
-  - **Parameters**:
-    - `name`: The name of the model to import.
-    - `folder`: The source folder.
-
-### Data Management
-
-- `export_cube(data, name, folder)`
-  - **Purpose**: Export data cubes to a specified folder in user workspace.
-  - **Parameters**:
-    - `data`: The data cube to export.
-    - `name`: The name to save the data cube under.
-    - `folder`: The destination folder.
-
-- `import_cube(name, folder)`
-  - **Purpose**: Import data cubes from a specified folder in user workspace.
-  - **Parameters**:
-    - `name`: The name of the data cube to import.
-    - `folder`: The source folder.
-
-### Results and Validation
-
-- `load_result(job_id)`
-  - **Purpose**: Load the result of a specified job.
-  - **Parameters**:
-    - `job_id`: The identifier for the job whose result is to be loaded.
-
-- `ml_label_class`
-  - **Purpose**: Label the classes in the modeling context.
-  - **Notes**: Use this function to assign labels to data classes post-prediction.
-
-- `ml_validate`
-  - **Purpose**: Validate models and predictions against the dataset.
-  - **Notes**: This function checks for accuracy and reliability of model outputs.
-
-## Examples
-
-TO DO
-
-
-## Naming Convention Explanation
-
-### `mlm_class_` and `mlm_regr_` Prefixes
-
-- **Model Definition**:
-  - The `mlm_class_` prefix is used for functions that define and initialize classification models. These functions create model structures but do not perform any training, returning untrained classification models.
-  - The `mlm_regr_` prefix applies to regression models. These functions provide the framework of a regression model without training, outputting untrained models ready for fitting with data.
-  - This setup allows users to configure the model architecture and parameters prior to training.
-
-### `ml_` Prefix
-
-- **General ML Operations**:
-  - The `ml_` prefix encompasses general machine learning operations applicable across stages, such as model training, prediction, and post-processing. These processes include key tasks like fitting (`ml_fit`), predicting (`ml_predict`), and validating (`ml_validate`).
-
-## Adopting openEO Conventions
-
-### `load_` Prefix
-
-- **Loading Resources**:
-  - The `load_` prefix is used for functions involved in retrieving or loading resources, conforming to openEO's approach to data acquisition.
-  - Examples:
-    - `load_ml_model(uri)`: Loads a machine learning model from storage using its URI.
-    - `load_result(job_id)`: Retrieves the result of a job given its ID.
-
-### `export_` Prefix
-
-- **Exporting Data/Models**:
-  - The `export_` prefix denotes functions that handle exporting data or models.
-  - Examples:
-    - `export_cube(data, name, folder)`: Exports a data cube to a designated folder.
-    - `export_model(model, name, folder)`: Exports a trained model for external use or storage.
-
-### `import_` Prefix
-
-- **Importing Resources**:
-  - The `import_` prefix indicates functions intended to bring data or models from external resources into the application.
-  - Examples:
-    - `import_cube(name, folder)`: Imports a data cube from a designated folder.
-    - `import_model(name, folder)`: Imports a model from a specified folder.
-
-## Advantages of Alignment with openEO
-
-- **Familiarity**: Leveraging openEO conventions helps users quickly understand and operate the API, especially those familiar with openEO.
-- **Consistency**: This reduces confusion by maintaining a uniform interaction model for data, models, and results.
-- **Integration**: Adhering to openEO standards facilitates easier integration with other tools and services within its ecosystem.
-
-This structured approach allows users to effectively transition from defining their models to executing a comprehensive ML workflow tailored for Earth Observation data cubes.
-		
\ No newline at end of file

From 3539d05c2438e320fc382218fc55ceb56d10a2cf Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Tue, 21 Oct 2025 15:42:45 +0200
Subject: [PATCH 47/52] ml proposals update

---
 CHANGELOG.md               |  1 -
 proposals/import_cube.json | 33 ---------------------------------
 2 files changed, 34 deletions(-)
 delete mode 100644 proposals/import_cube.json

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 7402db23..2db39a19 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -395,4 +395,3 @@ Older versions of the processes were released as part of the openEO API, see the
 [0.4.2]: <https://github.com/Open-EO/openeo-processes/compare/0.4.1...0.4.2>
 [0.4.1]: <https://github.com/Open-EO/openeo-processes/compare/0.4.0...0.4.1>
 [0.4.0]: <https://github.com/Open-EO/openeo-processes/tree/0.4.0>
-
diff --git a/proposals/import_cube.json b/proposals/import_cube.json
deleted file mode 100644
index 1591b65b..00000000
--- a/proposals/import_cube.json
+++ /dev/null
@@ -1,33 +0,0 @@
-{
-    "id": "import_cube",
-    "summary": "Import a data cube from a workspace",
-    "description": "Imports a previously exported data cube from a specified workspace location. This allows for reusing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) across different machine learning workflows.",
-    "categories": [
-        "Import",
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "name",
-            "description": "The name of the data cube to import. This should match the name used when the data cube was exported.",
-            "schema": {
-                "type": "string"
-            }
-        },
-        {
-            "name": "folder",
-            "description": "The folder path within the workspace where the data cube is stored.",
-            "schema": {
-                "type": "string"
-            }
-        }
-    ],
-    "returns": {
-        "description": "The imported data cube that can be used in subsequent processing steps.",
-        "schema": {
-            "type": "object",
-            "subtype": "datacube"
-        }
-    }
-} 
\ No newline at end of file

From 23c48e323f94af04ad0919e11eb6c56011134d86 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Tue, 21 Oct 2025 15:59:40 +0200
Subject: [PATCH 48/52] clean up

---
 .github/workflows/docs.yml                 |   2 -
 proposals-ml/export_cube.json              |  40 ------
 proposals-ml/import_cube.json              |  33 -----
 proposals-ml/load_ml_model.json            |  43 ------
 proposals-ml/load_stac_ml.json             |  74 ----------
 proposals-ml/ml_fit.json                   |  64 ---------
 proposals-ml/ml_label_class.json           |  38 -----
 proposals-ml/ml_predict.json               |  40 ------
 proposals-ml/ml_predict_probabilities.json |  39 -----
 proposals-ml/ml_smooth_class.json          |  66 ---------
 proposals-ml/ml_uncertainty_class.json     |  40 ------
 proposals-ml/mlm_class_catboost.json       |  62 --------
 proposals-ml/mlm_class_lighttae.json       | 128 -----------------
 proposals-ml/mlm_class_mlp.json            | 137 ------------------
 proposals-ml/mlm_class_random_forest.json  |  67 ---------
 proposals-ml/mlm_class_svm.json            | 112 ---------------
 proposals-ml/mlm_class_tae.json            | 128 -----------------
 proposals-ml/mlm_class_tempcnn.json        | 159 ---------------------
 proposals-ml/mlm_class_xgboost.json        |  89 ------------
 proposals-ml/mlm_regr_random_forest.json   |  68 ---------
 proposals-ml/mlm_regr_svm.json             | 122 ----------------
 proposals-ml/save_ml_model.json            |  74 ----------
 22 files changed, 1625 deletions(-)
 delete mode 100644 proposals-ml/export_cube.json
 delete mode 100644 proposals-ml/import_cube.json
 delete mode 100644 proposals-ml/load_ml_model.json
 delete mode 100644 proposals-ml/load_stac_ml.json
 delete mode 100644 proposals-ml/ml_fit.json
 delete mode 100644 proposals-ml/ml_label_class.json
 delete mode 100644 proposals-ml/ml_predict.json
 delete mode 100644 proposals-ml/ml_predict_probabilities.json
 delete mode 100644 proposals-ml/ml_smooth_class.json
 delete mode 100644 proposals-ml/ml_uncertainty_class.json
 delete mode 100644 proposals-ml/mlm_class_catboost.json
 delete mode 100644 proposals-ml/mlm_class_lighttae.json
 delete mode 100644 proposals-ml/mlm_class_mlp.json
 delete mode 100644 proposals-ml/mlm_class_random_forest.json
 delete mode 100644 proposals-ml/mlm_class_svm.json
 delete mode 100644 proposals-ml/mlm_class_tae.json
 delete mode 100644 proposals-ml/mlm_class_tempcnn.json
 delete mode 100644 proposals-ml/mlm_class_xgboost.json
 delete mode 100644 proposals-ml/mlm_regr_random_forest.json
 delete mode 100644 proposals-ml/mlm_regr_svm.json
 delete mode 100644 proposals-ml/save_ml_model.json

diff --git a/.github/workflows/docs.yml b/.github/workflows/docs.yml
index c8642243..ce7ba389 100644
--- a/.github/workflows/docs.yml
+++ b/.github/workflows/docs.yml
@@ -4,8 +4,6 @@ on:
     types: [published]
   push:
     branches:
-      - dev
-      - ml-poc
       - draft
       - master
 jobs:
diff --git a/proposals-ml/export_cube.json b/proposals-ml/export_cube.json
deleted file mode 100644
index 8810c653..00000000
--- a/proposals-ml/export_cube.json
+++ /dev/null
@@ -1,40 +0,0 @@
-{
-    "id": "export_cube",
-    "summary": "Export a data cube to a workspace",
-    "description": "Exports a processed data cube to a specified workspace location. This allows for storing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) for later use across various machine learning models.\n\nThe workspace is recommended to be on the same server as the implementation, this ensures better performance.",
-    "categories": [
-        "Export",
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "data",
-            "description": "The data cube to be exported. This should be a processed data cube that you want to store for later use.",
-            "schema": {
-                "type": "object",
-                "subtype": "datacube"
-            }
-        },
-        {
-            "name": "name",
-            "description": "The name to be given to the exported data cube. This will be used to identify the data when importing it later.",
-            "schema": {
-                "type": "string"
-            }
-        },
-        {
-            "name": "folder",
-            "description": "The folder path within the workspace where the data cube should be stored.",
-            "schema": {
-                "type": "string"
-            }
-        }
-    ],
-    "returns": {
-        "description": "A boolean indicating whether the export was successful.",
-        "schema": {
-            "type": "boolean"
-        }
-    }
-} 
\ No newline at end of file
diff --git a/proposals-ml/import_cube.json b/proposals-ml/import_cube.json
deleted file mode 100644
index 1591b65b..00000000
--- a/proposals-ml/import_cube.json
+++ /dev/null
@@ -1,33 +0,0 @@
-{
-    "id": "import_cube",
-    "summary": "Import a data cube from a workspace",
-    "description": "Imports a previously exported data cube from a specified workspace location. This allows for reusing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) across different machine learning workflows.",
-    "categories": [
-        "Import",
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "name",
-            "description": "The name of the data cube to import. This should match the name used when the data cube was exported.",
-            "schema": {
-                "type": "string"
-            }
-        },
-        {
-            "name": "folder",
-            "description": "The folder path within the workspace where the data cube is stored.",
-            "schema": {
-                "type": "string"
-            }
-        }
-    ],
-    "returns": {
-        "description": "The imported data cube that can be used in subsequent processing steps.",
-        "schema": {
-            "type": "object",
-            "subtype": "datacube"
-        }
-    }
-} 
\ No newline at end of file
diff --git a/proposals-ml/load_ml_model.json b/proposals-ml/load_ml_model.json
deleted file mode 100644
index 8980a20d..00000000
--- a/proposals-ml/load_ml_model.json
+++ /dev/null
@@ -1,43 +0,0 @@
-{
-    "id": "load_ml_model",
-    "summary": "Load a machine learning model by ID",
-    "description": "Loads a machine learning model that is managed by the current back-end, identified by a back-end specific model identifier.\n\nThis allows the back-end to host and optimize models for efficient inference and training (e.g., via model-specific libraries such as Terratorch). Back-end specific models can be referenced by a string identifier similar to collections.\n\nIf you want to load a model from a STAC Item implementing the `mlm` extension, use `load_stac_ml()` instead.",
-    "categories": [
-        "machine learning",
-        "import"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "id",
-            "description": "The back-end specific identifier of the machine learning model to load.",
-            "schema": {
-                "type": "string",
-                "pattern": "^[\\w\\-\\.~/]+$"
-            }
-        }
-    ],
-    "returns": {
-        "description": "A machine learning model to be used with machine learning processes such as `ml_predict()`.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://github.com/stac-extensions/mlm",
-            "title": "Machine Learning Model STAC extension",
-            "type": "text/html",
-            "rel": "about"
-        },
-        {
-            "href": "https://openeo.org/documentation/1.0/",
-            "title": "openEO API documentation",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-}
-
-
diff --git a/proposals-ml/load_stac_ml.json b/proposals-ml/load_stac_ml.json
deleted file mode 100644
index f776bc06..00000000
--- a/proposals-ml/load_stac_ml.json
+++ /dev/null
@@ -1,74 +0,0 @@
-{
-    "id": "load_stac_ml",
-    "summary": "Load a ML model from a STAC Item",
-    "description": "Loads a machine learning model from a STAC Item.\n\nSuch a model could be trained and saved as part of a previous batch job with processes such as  ``ml_fit()`` and ``save_ml_model()`` or  externally hosted models.",
-    "categories": [
-        "machine learning",
-        "import"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "uri",
-            "description": "The STAC Item to load the machine learning model from. The STAC Item must implement the [`mlm`](https://github.com/stac-extensions/mlm) extension. This parameter can point to a remote STAC Item via ``URL`` or a local JSON file.",
-            "schema": [
-                {
-                    "title": "URL",
-                    "type": "string",
-                    "format": "uri",
-                    "subtype": "uri",
-                    "pattern": "^https?://"
-                },
-                {
-                    "title": "User-uploaded File",
-                    "type": "string",
-                    "subtype": "file-path",
-                    "pattern": "^[^\r\n\\:'\"]+$"
-                }
-            ]
-        },
-        {
-            "name": "model_asset",
-            "description": "The Asset name of the given STAC Item which represents the actual ML model. The asset must list ``mlm:model`` as its role. If only one asset lists ``mlm:model`` as its role, this parameter is optional as this asset will be used by default. If multiple assets list ``mlm:model`` as their role, this parameter is required to determine which asset to use.",
-            "schema": {
-                "type": "string"
-            },
-            "default": null,
-            "optional": true
-        },
-        {
-            "name": "input_index",
-            "description": "STAC:MLM items supports multiple ML model input specification. This parameter specifies the index of the input specification in the ``mlm:input`` array to use for prediction or training. As ``mlm:input`` is an array, the first input in the array has index 0.",
-            "schema": {
-                "type": "integer"
-            },
-            "default": 0,
-            "optional": true
-        },
-        {
-            "name": "output_index",
-            "description": "STAC:MLM items supports multiple ML model output specification. This parameter specifies the index of the output specification in the ``mlm:output`` array to use for prediction or training. As ``mlm:output`` is an array, the first output in the array has index 0.",
-            "schema": {
-                "type": "integer"
-            },
-            "default": 0,
-            "optional": true
-        }
-    ],
-    "returns": {
-        "description": "A machine learning model to be used with machine learning processes such as ``ml_predict()``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://github.com/stac-extensions/mlm",
-            "title": "Machine Learning Model STAC extension",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-}
-
diff --git a/proposals-ml/ml_fit.json b/proposals-ml/ml_fit.json
deleted file mode 100644
index 3825773c..00000000
--- a/proposals-ml/ml_fit.json
+++ /dev/null
@@ -1,64 +0,0 @@
-{
-    "id": "ml_fit",
-    "summary": "Train a machine learning model",
-    "description": "Executes the fit of a specified machine learning model based on training data.\n\nThe function is generic and supports different machine learning models.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "model",
-            "description": "The machine learning model to be trained. This should be an instance of a model that supports the `ml_fit` method.",
-            "schema": {
-                "type": "object",
-                "subtype": "ml-model"
-            }
-        },
-        {
-            "name": "training_set",
-            "description": "The training set for the model, provided as a vector data cube. This set contains both the independent variables and the dependent variable that the model analyzes to learn patterns and relationships within the data.",
-            "schema": [
-                {
-                    "type": "object",
-                    "subtype": "datacube",
-                    "dimensions": [
-                        {
-                            "type": "geometry"
-                        },
-                        {
-                            "type": "bands"
-                        }
-                    ]
-                },
-                {
-                    "type": "object",
-                    "subtype": "datacube",
-                    "dimensions": [
-                        {
-                            "type": "geometry"
-                        },
-                        {
-                            "type": "other"
-                        }
-                    ]
-                }
-            ]
-        },
-        {
-            "name": "target",
-            "description": "The name of the variable in the training set that serves as the target for model training.",
-            "schema": {
-                "type": "string"
-            }
-        }
-    ],
-    "returns": {
-        "description": "A trained model object that can be saved with `save_ml_model()` and restored with `load_ml_model()`.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    }
-}
-
diff --git a/proposals-ml/ml_label_class.json b/proposals-ml/ml_label_class.json
deleted file mode 100644
index 649fe0e2..00000000
--- a/proposals-ml/ml_label_class.json
+++ /dev/null
@@ -1,38 +0,0 @@
-{
-    "id": "ml_label_class",
-    "summary": "Convert probability data cube to labeled data cube",
-    "description": "Converts a probability data cube to a labeled data cube by applying softmax normalization and selecting the class with the highest probability. Optionally allows mapping of class indices to custom labels.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "data",
-            "description": "The input probability data cube to be labeled. Each band should represent the probability of a different class.",
-            "schema": {
-                "type": "object",
-                "subtype": "datacube"
-            }
-        },
-        {
-            "name": "labels",
-            "description": "Optional dictionary mapping class indices to custom labels. The dictionary keys are the class indices (as integers) and the values are the custom labels for each class. If not provided, the class indices will be used as labels.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "object",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A labeled data cube where each pixel contains the class label with the highest probability after softmax normalization.",
-        "schema": {
-            "type": "object",
-            "subtype": "datacube"
-        }
-    }
-} 
\ No newline at end of file
diff --git a/proposals-ml/ml_predict.json b/proposals-ml/ml_predict.json
deleted file mode 100644
index 4741bcd9..00000000
--- a/proposals-ml/ml_predict.json
+++ /dev/null
@@ -1,40 +0,0 @@
-{
-    "id": "ml_predict",
-    "summary": "Predict using ML",
-    "description": "Applies a machine learning model to a data cube of input features and returns the predicted values.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "data",
-            "description": "The data cube containing the input features.",
-            "schema": {
-                "type": "object",
-                "subtype": "datacube"
-            }
-        },
-        {
-            "name": "model",
-            "description": "A ML model that was trained.",
-            "schema": {
-                "type": "object",
-                "subtype": "ml-model"
-            }
-        }
-    ],
-    "returns": {
-        "description": "A data cube with the predicted values. It removes the specified dimensions and adds new dimension for the predicted values. It has the name `predictions` and is of type `other`. If a single value is returned, the dimension has a single label with name `0`.",
-        "schema": {
-            "type": "object",
-            "subtype": "datacube",
-            "dimensions": [
-                {
-                    "type": "other"
-                }
-            ]
-        }
-    }
-}
-
diff --git a/proposals-ml/ml_predict_probabilities.json b/proposals-ml/ml_predict_probabilities.json
deleted file mode 100644
index c44689b6..00000000
--- a/proposals-ml/ml_predict_probabilities.json
+++ /dev/null
@@ -1,39 +0,0 @@
-{
-    "id": "ml_predict_probabilities",
-    "summary": "Predict class probabilities using ML",
-    "description": "Applies a machine learning model to a data cube of input features and returns the predicted class probabilities.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "data",
-            "description": "The data cube containing the input features.",
-            "schema": {
-                "type": "object",
-                "subtype": "datacube"
-            }
-        },
-        {
-            "name": "model",
-            "description": "A ML model that was trained and supports probability predictions.",
-            "schema": {
-                "type": "object",
-                "subtype": "ml-model"
-            }
-        }
-    ],
-    "returns": {
-        "description": "A data cube with the predicted class probabilities. It removes the specified dimensions and adds a new dimension for the class probabilities. The dimension has the name `classes` and is of type `other`. Each label in the dimension represents a class, and the values are the probabilities for each class.",
-        "schema": {
-            "type": "object",
-            "subtype": "datacube",
-            "dimensions": [
-                {
-                    "type": "other"
-                }
-            ]
-        }
-    }
-}
\ No newline at end of file
diff --git a/proposals-ml/ml_smooth_class.json b/proposals-ml/ml_smooth_class.json
deleted file mode 100644
index 380de6ad..00000000
--- a/proposals-ml/ml_smooth_class.json
+++ /dev/null
@@ -1,66 +0,0 @@
-{
-    "id": "ml_smooth_class",
-    "summary": "Apply Bayesian smoothing to classification results",
-    "description": "Applies spatial smoothing to classification probabilities results from a machine learning model using Bayesian inference. This process helps reduce noise and improve classification accuracy by considering spatial context and neighbourhood information. The smoothing is based on a window-based approach.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "data",
-            "description": "The input classification probabilities data to be smoothed. This should be a classified data cube.",
-            "schema": {
-                "type": "object",
-                "subtype": "datacube"
-            }
-        },
-        {
-            "name": "window_size",
-            "description": "The size of the moving window used for spatial smoothing. Must be an odd number to ensure a central pixel.",
-            "optional": true,
-            "default": 7,
-            "schema": {
-                "type": "integer",
-                "minimum": 3,
-                "multipleOf": 2
-            }
-        },
-        {
-            "name": "neighborhood_fraction",
-            "description": "The fraction of neighbouring pixels to consider in the smoothing process. Controls the influence of local spatial information.",
-            "optional": true,
-            "default": 0.5,
-            "schema": {
-                "type": "number",
-                "minimum": 0,
-                "maximum": 1
-            }
-        },
-        {
-            "name": "smoothness",
-            "description": "Controls the degree of smoothing applied. Higher values result in smoother outputs but may lose fine details.",
-            "optional": true,
-            "default": 10,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        }
-    ],
-    "returns": {
-        "description": "A smoothed classification result with reduced noise and improved spatial consistency.",
-        "schema": {
-            "type": "object",
-            "subtype": "datacube"
-        }
-    },
-    "links": [
-        {
-            "href": "https://doi.org/10.3390/rs16234572",
-            "title": "Camara et al. (2024): Bayesian Inference for Post-Processing of Remote-Sensing Image Classification",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-} 
\ No newline at end of file
diff --git a/proposals-ml/ml_uncertainty_class.json b/proposals-ml/ml_uncertainty_class.json
deleted file mode 100644
index 86402906..00000000
--- a/proposals-ml/ml_uncertainty_class.json
+++ /dev/null
@@ -1,40 +0,0 @@
-{
-    "id": "ml_uncertainty_class",
-    "summary": "Estimate uncertainty of classification probabilities",
-    "description": "Calculates uncertainty estimates for classification probabilities using various methods. The uncertainty is derived from the probability distribution of class predictions.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "data",
-            "description": "The data cube containing the classification probabilities.",
-            "schema": {
-                "type": "object",
-                "subtype": "datacube"
-            }
-        },
-        {
-            "name": "method",
-            "description": "The method to use for uncertainty estimation. 'margin' calculates the difference between the two highest probabilities, 'ratio' uses the ratio of the two highest probabilities, and 'least-confidence' uses 1 minus the highest probability.",
-            "default": "margin",
-            "optional": true,
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "margin",
-                    "ratio",
-                    "least-confidence"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A data cube with the uncertainty estimates. The output maintains the same dimensions as the input data cube, but replaces the class probabilities with uncertainty values. Higher values indicate higher uncertainty in the classification.",
-        "schema": {
-            "type": "object",
-            "subtype": "datacube"
-        }
-    }
-} 
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_catboost.json b/proposals-ml/mlm_class_catboost.json
deleted file mode 100644
index 7a46d0a1..00000000
--- a/proposals-ml/mlm_class_catboost.json
+++ /dev/null
@@ -1,62 +0,0 @@
-{
-    "id": "mlm_class_catboost",
-    "summary": "Initialize a CatBoost classification model",
-    "description": "Initializes a CatBoost classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "iterations",
-            "description": "The maximum number of trees that can be built during the training process.",
-            "optional": true,
-            "default": 5,
-            "schema": {
-                "type": "integer",
-                "minimum": 1,
-                "maximum": 500
-            }
-        },
-        {
-            "name": "depth",
-            "description": "Depth of the trees in the CatBoost model.",
-            "optional": true,
-            "default": 5,
-            "schema": {
-                "type": "integer",
-                "minimum": 1,
-                "maximum": 16
-            }
-        },
-        {
-            "name": "seed",
-            "description": "The random seed used for training, for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": 0,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ],
-                "minimum": 0,
-                "maximum": 2147483647
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://catboost.ai/",
-            "title": "CatBoost Documentation",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-} 
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_lighttae.json b/proposals-ml/mlm_class_lighttae.json
deleted file mode 100644
index b9f88533..00000000
--- a/proposals-ml/mlm_class_lighttae.json
+++ /dev/null
@@ -1,128 +0,0 @@
-{
-    "id": "mlm_class_lighttae",
-    "summary": "Initialize a LightTAE classification model",
-    "description": "Initializes a Lightweight Temporal Self-Attention (LightTAE) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "epochs",
-            "description": "Number of training epochs.",
-            "optional": true,
-            "default": 150,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "batch_size",
-            "description": "Size of the training batches.",
-            "optional": true,
-            "default": 128,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "optimizer",
-            "description": "The optimizer to use for training.",
-            "optional": true,
-            "default": "adam",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "adam",
-                    "adabound",
-                    "adabelief",
-                    "madagrad",
-                    "nadam",
-                    "qhadam",
-                    "radam",
-                    "swats",
-                    "yogi"
-                ]
-            }
-        },
-        {
-            "name": "learning_rate",
-            "description": "The learning rate for the optimizer.",
-            "optional": true,
-            "default": 0.0005,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "epsilon",
-            "description": "A small constant for numerical stability in the optimizer.",
-            "optional": true,
-            "default": 1e-8,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "weight_decay",
-            "description": "Weight decay (L2 penalty) for the optimizer.",
-            "optional": true,
-            "default": 0.0007,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "lr_decay_epochs",
-            "description": "Number of epochs between learning rate decay steps.",
-            "optional": true,
-            "default": 50,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "lr_decay_rate",
-            "description": "Rate at which the learning rate decays.",
-            "optional": true,
-            "default": 1,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://doi.org/10.1007/978-3-030-65742-0_12",
-            "title": "V. S. F. Garnot and L. Landrieu (2020), Lightweight Temporal Self-attention for Classifying Satellite Images Time Series",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-} 
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_mlp.json b/proposals-ml/mlm_class_mlp.json
deleted file mode 100644
index 45722c74..00000000
--- a/proposals-ml/mlm_class_mlp.json
+++ /dev/null
@@ -1,137 +0,0 @@
-{
-    "id": "mlm_class_mlp",
-    "summary": "Initialize an MLP classification model",
-    "description": "Initializes a Multi-Layer Perceptron (MLP) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "hidden_layer_sizes",
-            "description": "List of integers specifying the number of neurons in each hidden layer.",
-            "default": [
-                512,
-                512,
-                512
-            ],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 1
-            }
-        },
-        {
-            "name": "activation",
-            "description": "Activation function for the hidden layers.",
-            "optional": true,
-            "default": "relu",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "relu",
-                    "tanh",
-                    "logistic"
-                ]
-            }
-        },
-        {
-            "name": "dropout_rates",
-            "description": "List of numbers between 0 and 1 specifying the dropout rate for each hidden layer.",
-            "default": [
-                0.4,
-                0.3,
-                0.2
-            ],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "number",
-                    "minimum": 0,
-                    "maximum": 1
-                },
-                "minItems": 1
-            }
-        },
-        {
-            "name": "epochs",
-            "description": "Number of training epochs.",
-            "optional": true,
-            "default": 100,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "batch_size",
-            "description": "Size of the training batches.",
-            "optional": true,
-            "default": 64,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "optimizer",
-            "description": "The optimizer to use for training.",
-            "optional": true,
-            "default": "adam",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "adam",
-                    "adabound",
-                    "adabelief",
-                    "madagrad",
-                    "nadam",
-                    "qhadam",
-                    "radam",
-                    "swats",
-                    "yogi"
-                ]
-            }
-        },
-        {
-            "name": "learning_rate",
-            "description": "The learning rate for the optimizer.",
-            "optional": true,
-            "default": 0.001,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://arxiv.org/abs/1611.06455",
-            "title": "Z. Wang, W. Yan, and T. Oates (2017), Time Series Classification from Scratch with Deep Neural Networks: A Strong Baseline",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-}
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_random_forest.json b/proposals-ml/mlm_class_random_forest.json
deleted file mode 100644
index 0881e0f0..00000000
--- a/proposals-ml/mlm_class_random_forest.json
+++ /dev/null
@@ -1,67 +0,0 @@
-{
-    "id": "mlm_class_random_forest",
-    "summary": "Initialize a random forest classification model",
-    "description": "Initializes a Random Forest classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "max_variables",
-            "description": "Specifies how many split variables will be used at a node.\n\nThe following options are available:\n\n- *integer*: The given number of variables are considered for each split.\n- `all`: All variables are considered for each split.\n- `log2`: The logarithm with base 2 of the number of variables are considered for each split.\n- `onethird`: A third of the number of variables are considered for each split.\n- `sqrt`: The square root of the number of variables are considered for each split. This is often the default for classification.",
-            "schema": [
-                {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                {
-                    "type": "string",
-                    "enum": [
-                        "all",
-                        "log2",
-                        "onethird",
-                        "sqrt"
-                    ]
-                }
-            ]
-        },
-        {
-            "name": "num_trees",
-            "description": "The number of trees build within the Random Forest classification.",
-            "optional": true,
-            "default": 100,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for the random sampling in training. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://doi.org/10.1023/A:1010933404324",
-            "title": "Breiman (2001): Random Forests",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-}
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_svm.json b/proposals-ml/mlm_class_svm.json
deleted file mode 100644
index bf0a8ca7..00000000
--- a/proposals-ml/mlm_class_svm.json
+++ /dev/null
@@ -1,112 +0,0 @@
-{
-    "id": "mlm_class_svm",
-    "summary": "Initialize an SVM classification model",
-    "description": "Initializes a Support Vector Machine (SVM) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "kernel",
-            "description": "Specifies the kernel type to be used in the algorithm.",
-            "optional": true,
-            "default": "rbf",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "linear",
-                    "poly",
-                    "rbf",
-                    "sigmoid"
-                ]
-            }
-        },
-        {
-            "name": "C",
-            "description": "Regularization parameter. The strength of the regularization is inversely proportional to C. Must be strictly positive.",
-            "optional": true,
-            "default": 1,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "gamma",
-            "description": "Kernel coefficient for 'rbf', 'poly', and 'sigmoid'. Higher values lead to tighter fits.",
-            "optional": true,
-            "default": 1,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "degree",
-            "description": "Degree of the polynomial kernel function (only relevant for 'poly' kernel).",
-            "optional": true,
-            "default": 3,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "coef0",
-            "description": "Independent term in the kernel function (only relevant for 'poly' and 'sigmoid' kernels).",
-            "optional": true,
-            "default": 0,
-            "schema": {
-                "type": "number"
-            }
-        },
-        {
-            "name": "tolerance",
-            "description": "Tolerance of termination criterion.",
-            "optional": true,
-            "default": 0.001,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "cachesize",
-            "description": "Size of the kernel cache in MB.",
-            "optional": true,
-            "default": 1000,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://link.springer.com/article/10.1007/BF00994018",
-            "title": "C. Cortes and V. Vapnik (1995), Support-vector networks",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-}
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_tae.json b/proposals-ml/mlm_class_tae.json
deleted file mode 100644
index 6f81e30e..00000000
--- a/proposals-ml/mlm_class_tae.json
+++ /dev/null
@@ -1,128 +0,0 @@
-{
-    "id": "mlm_class_tae",
-    "summary": "Initialize a TAE classification model",
-    "description": "Initializes a Temporal Self-Attention (TAE) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "epochs",
-            "description": "Number of training epochs.",
-            "optional": true,
-            "default": 150,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "batch_size",
-            "description": "Size of the training batches.",
-            "optional": true,
-            "default": 64,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "optimizer",
-            "description": "The optimizer to use for training.",
-            "optional": true,
-            "default": "adam",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "adam",
-                    "adabound",
-                    "adabelief",
-                    "madagrad",
-                    "nadam",
-                    "qhadam",
-                    "radam",
-                    "swats",
-                    "yogi"
-                ]
-            }
-        },
-        {
-            "name": "learning_rate",
-            "description": "The learning rate for the optimizer.",
-            "optional": true,
-            "default": 0.001,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "epsilon",
-            "description": "A small constant for numerical stability in the optimizer.",
-            "optional": true,
-            "default": 1e-8,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "weight_decay",
-            "description": "Weight decay (L2 penalty) for the optimizer.",
-            "optional": true,
-            "default": 0.000001,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "lr_decay_epochs",
-            "description": "Number of epochs between learning rate decay steps.",
-            "optional": true,
-            "default": 1,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "lr_decay_rate",
-            "description": "Rate at which the learning rate decays.",
-            "optional": true,
-            "default": 0.95,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://doi.org/10.1109/CVPR42600.2020.01234",
-            "title": "V. Garnot, L. Landrieu, S. Giordano, and N. Chehata (2020), Satellite Image Time Series Classification With Pixel-Set Encoders and Temporal Self-Attention",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-} 
\ No newline at end of file
diff --git a/proposals-ml/mlm_class_tempcnn.json b/proposals-ml/mlm_class_tempcnn.json
deleted file mode 100644
index cc9da6f6..00000000
--- a/proposals-ml/mlm_class_tempcnn.json
+++ /dev/null
@@ -1,159 +0,0 @@
-{
-    "id": "mlm_class_tempcnn",
-    "summary": "Initialize a TempCNN classification model",
-    "description": "Initializes a Temporal Convolutional Neural Network (TempCNN) classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "cnn_layers",
-            "description": "List of integers specifying the number of filters in each convolutional layer.",
-            "default": [
-                256,
-                256,
-                256
-            ],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 1
-            }
-        },
-        {
-            "name": "cnn_kernels",
-            "description": "List of integers specifying the kernel size for each convolutional layer.",
-            "default": [
-                7,
-                7,
-                7
-            ],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 1
-            }
-        },
-        {
-            "name": "cnn_dropout_rates",
-            "description": "List of numbers between 0 and 1 specifying the dropout rate for each convolutional layer.",
-            "default": [
-                0.2,
-                0.2,
-                0.2
-            ],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "number",
-                    "minimum": 0,
-                    "maximum": 1
-                },
-                "minItems": 1
-            }
-        },
-        {
-            "name": "dense_layer_nodes",
-            "description": "Number of nodes in the dense layer.",
-            "default": 256,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "dense_layer_dropout_rate",
-            "description": "Dropout rate for the dense layer.",
-            "default": 0.5,
-            "schema": {
-                "type": "number",
-                "minimum": 0,
-                "maximum": 1
-            }
-        },
-        {
-            "name": "epochs",
-            "description": "Number of training epochs.",
-            "optional": true,
-            "default": 100,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "batch_size",
-            "description": "Size of the training batches.",
-            "optional": true,
-            "default": 64,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "optimizer",
-            "description": "The optimizer to use for training.",
-            "optional": true,
-            "default": "adam",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "adam",
-                    "adabound",
-                    "adabelief",
-                    "madagrad",
-                    "nadam",
-                    "qhadam",
-                    "radam",
-                    "swats",
-                    "yogi"
-                ]
-            }
-        },
-        {
-            "name": "learning_rate",
-            "description": "The learning rate for the optimizer.",
-            "optional": true,
-            "default": 0.001,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://doi.org/10.3390/rs11050523",
-            "title": "Pelletier et al. (2019): Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-}
diff --git a/proposals-ml/mlm_class_xgboost.json b/proposals-ml/mlm_class_xgboost.json
deleted file mode 100644
index a7b21f50..00000000
--- a/proposals-ml/mlm_class_xgboost.json
+++ /dev/null
@@ -1,89 +0,0 @@
-{
-    "id": "mlm_class_xgboost",
-    "summary": "Initialize an XGBoost classification model",
-    "description": "Initializes an XGBoost classification model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "learning_rate",
-            "description": "Step size shrinkage used in update to prevent overfitting.",
-            "optional": true,
-            "default": 0.15,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "max_depth",
-            "description": "Maximum depth of a tree.",
-            "optional": true,
-            "default": 5,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "min_child_weight",
-            "description": "Minimum sum of instance weight (hessian) needed in a child.",
-            "optional": true,
-            "default": 1,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "subsample",
-            "description": "Subsample ratio of the training instance.",
-            "optional": true,
-            "default": 0.8,
-            "schema": {
-                "type": "number",
-                "minimum": 0,
-                "maximum": 1
-            }
-        },
-        {
-            "name": "min_split_loss",
-            "description": "Minimum loss reduction required to make a further partition on a leaf node of the tree.",
-            "optional": true,
-            "default": 1,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://dl.acm.org/doi/10.1145/2939672.2939785",
-            "title": "Chen and Guestrin (2016), XGBoost: A Scalable Tree Boosting System",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-}
\ No newline at end of file
diff --git a/proposals-ml/mlm_regr_random_forest.json b/proposals-ml/mlm_regr_random_forest.json
deleted file mode 100644
index f643a73a..00000000
--- a/proposals-ml/mlm_regr_random_forest.json
+++ /dev/null
@@ -1,68 +0,0 @@
-{
-    "id": "mlm_regr_random_forest",
-    "summary": "Initialize a random forest regression model",
-    "description": "Initializes a Random Forest regression model. This component prepares the model structure but does not perform training. The model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "max_variables",
-            "description": "Specifies how many split variables will be used at a node.\n\nThe following options are available:\n\n- *integer*: The given number of variables are considered for each split.\n- `all`: All variables are considered for each split.\n- `log2`: The logarithm with base 2 of the number of variables are considered for each split.\n- `onethird`: A third of the number of variables are considered for each split. This is often the default for regression.\n- `sqrt`: The square root of the number of variables are considered for each split.",
-            "schema": [
-                {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                {
-                    "type": "string",
-                    "enum": [
-                        "all",
-                        "log2",
-                        "onethird",
-                        "sqrt"
-                    ]
-                }
-            ]
-        },
-        {
-            "name": "num_trees",
-            "description": "The number of trees build within the Random Forest regression.",
-            "optional": true,
-            "default": 100,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for the random sampling in training. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://doi.org/10.1023/A:1010933404324",
-            "title": "Breiman (2001): Random Forests",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-}
-
diff --git a/proposals-ml/mlm_regr_svm.json b/proposals-ml/mlm_regr_svm.json
deleted file mode 100644
index ba63930c..00000000
--- a/proposals-ml/mlm_regr_svm.json
+++ /dev/null
@@ -1,122 +0,0 @@
-{
-    "id": "mlm_regr_svm",
-    "summary": "Initialize an SVM regression model",
-    "description": "Initializes a Support Vector Machine (SVM) regression model. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "kernel",
-            "description": "Specifies the kernel type to be used in the algorithm.",
-            "optional": true,
-            "default": "rbf",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "linear",
-                    "poly",
-                    "rbf",
-                    "sigmoid"
-                ]
-            }
-        },
-        {
-            "name": "C",
-            "description": "Regularization parameter. The strength of the regularization is inversely proportional to C. Must be strictly positive.",
-            "optional": true,
-            "default": 1,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "epsilon",
-            "description": "Epsilon in the epsilon-SVR model. Specifies the epsilon-tube within which no penalty is associated in the training loss function with points predicted within a distance epsilon from the actual value.",
-            "optional": true,
-            "default": 0.1,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "gamma",
-            "description": "Kernel coefficient for 'rbf', 'poly', and 'sigmoid'. Higher values lead to tighter fits.",
-            "optional": true,
-            "default": 1,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "degree",
-            "description": "Degree of the polynomial kernel function (only relevant for 'poly' kernel).",
-            "optional": true,
-            "default": 3,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "coef0",
-            "description": "Independent term in the kernel function (only relevant for 'poly' and 'sigmoid' kernels).",
-            "optional": true,
-            "default": 0,
-            "schema": {
-                "type": "number"
-            }
-        },
-        {
-            "name": "tolerance",
-            "description": "Tolerance of termination criterion.",
-            "optional": true,
-            "default": 0.001,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "cachesize",
-            "description": "Size of the kernel cache in MB.",
-            "optional": true,
-            "default": 1000,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://link.springer.com/chapter/10.1007/978-1-4302-5990-9_4",
-            "title": "Awad, M., Khanna, R., Awad, M., & Khanna, R. (2015). Support vector regression. Efficient learning machines: Theories, concepts, and applications for engineers and system designers.",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-} 
\ No newline at end of file
diff --git a/proposals-ml/save_ml_model.json b/proposals-ml/save_ml_model.json
deleted file mode 100644
index af593363..00000000
--- a/proposals-ml/save_ml_model.json
+++ /dev/null
@@ -1,74 +0,0 @@
-{
-    "id": "save_ml_model",
-    "summary": "Save a ML model",
-    "description": "Saves a machine learning model as part of a batch job.\n\nThe model will be accompanied by a separate STAC Item that implements the [mlm-model extension](https://github.com/stac-extensions/mlm).",
-    "categories": [
-        "machine learning",
-        "import"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "data",
-            "description": "The data to store as a machine learning model.",
-            "schema": {
-                "type": "object",
-                "subtype": "ml-model"
-            }
-        },
-        {
-            "name": "name",
-            "description": "A distinct name of the model.",
-            "schema": {
-                "type": "string"
-            }
-        },
-        {
-            "name": "tasks",
-            "description": "Specifies the ML tasks which the model can be used for.",
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "string",
-                    "enum": [
-                        "regression",
-                        "classification",
-                        "object-detection",
-                        "detection",
-                        "scene-classification",
-                        "segmentation",
-                        "semantic-segmentation",
-                        "similarity-search",
-                        "generative",
-                        "image-captioning",
-                        "super-resolution"
-                    ]
-                }
-            }
-        },
-        {
-            "name": "options",
-            "description": "Additional parameters to create the file(s).",
-            "schema": {
-                "type": "object",
-                "additionalParameters": false
-            },
-            "default": {},
-            "optional": true
-        }
-    ],
-    "returns": {
-        "description": "Returns `false` if the process failed to store the model, `true` otherwise.",
-        "schema": {
-            "type": "boolean"
-        }
-    },
-    "links": [
-        {
-            "href": "https://github.com/stac-extensions/mlm",
-            "title": "Machine Learning Model STAC extension",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-}
\ No newline at end of file

From e88ea7fd3e16820ebd1b657193e4bb0b1ed5353a Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Tue, 21 Oct 2025 18:04:12 +0200
Subject: [PATCH 49/52] remove not needed poc

---
 proposals-ml/mlm_class_1dcnn.json | 178 ------------------------------
 proposals/export_cube.json        |  40 -------
 2 files changed, 218 deletions(-)
 delete mode 100644 proposals-ml/mlm_class_1dcnn.json
 delete mode 100644 proposals/export_cube.json

diff --git a/proposals-ml/mlm_class_1dcnn.json b/proposals-ml/mlm_class_1dcnn.json
deleted file mode 100644
index 887276c0..00000000
--- a/proposals-ml/mlm_class_1dcnn.json
+++ /dev/null
@@ -1,178 +0,0 @@
-{
-    "id": "mlm_class_1dcnn",
-    "summary": "Initialize a 1D CNN classification model",
-    "description": "Initializes a 1D Convolutional Neural Network (CNN) classification model. The number of input channels and output classes are inferred automatically from the training data at fit time. This component sets up the model structure but does not perform training or handle data splitting. The resulting model can be trained later using ``ml_fit``.",
-    "categories": [
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "conv_filters",
-            "description": "List of integers specifying the number of filters in each convolutional layer. The final output layer for classification will be added automatically based on the number of classes in the training data.",
-            "default": [64, 128, 256, 512],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "conv_kernels",
-            "description": "List of integers or 'global' for kernel size in each convolutional layer. The 4th conv uses a global kernel (covers the entire sequence at that layer). The final output layer for classification will use a 1x1 kernel and be added automatically.",
-            "default": [3, 3, 3, "global"],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "oneOf": [
-                        {"type": "integer", "minimum": 1},
-                        {"type": "string", "enum": ["global"]}
-                    ]
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "conv_strides",
-            "description": "List of integers specifying the stride for each convolutional layer. The final output layer for classification will use stride 1 and be added automatically.",
-            "default": [1, 1, 1, 1],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "use_batchnorm",
-            "description": "List of booleans specifying whether to use batch normalization after each convolutional layer. The final output layer will not use batch normalization.",
-            "default": [true, true, true, false],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "boolean"
-                },
-                "minItems": 4,
-                "maxItems": 4
-            }
-        },
-        {
-            "name": "activation",
-            "description": "Activation function to use after each convolutional layer.",
-            "default": "relu",
-            "schema": {
-                "type": "string",
-                "enum": ["relu", "tanh", "sigmoid", "leakyrelu"]
-            }
-        },
-        {
-            "name": "maxpool_sizes",
-            "description": "List of integers specifying the pool size for each max pooling layer (after each of the first 3 conv blocks).",
-            "default": [2, 2, 2],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 3,
-                "maxItems": 3
-            }
-        },
-        {
-            "name": "maxpool_strides",
-            "description": "List of integers specifying the stride for each max pooling layer.",
-            "default": [2, 2, 2],
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "integer",
-                    "minimum": 1
-                },
-                "minItems": 3,
-                "maxItems": 3
-            }
-        },
-        {
-            "name": "epochs",
-            "description": "Number of training epochs.",
-            "optional": true,
-            "default": 100,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "batch_size",
-            "description": "Size of the training batches.",
-            "optional": true,
-            "default": 64,
-            "schema": {
-                "type": "integer",
-                "minimum": 1
-            }
-        },
-        {
-            "name": "optimizer",
-            "description": "The optimizer to use for training.",
-            "optional": true,
-            "default": "adam",
-            "schema": {
-                "type": "string",
-                "enum": [
-                    "adam",
-                    "sgd",
-                    "rmsprop",
-                    "adagrad",
-                    "nadam"
-                ]
-            }
-        },
-        {
-            "name": "learning_rate",
-            "description": "The learning rate for the optimizer.",
-            "optional": true,
-            "default": 0.001,
-            "schema": {
-                "type": "number",
-                "minimum": 0
-            }
-        },
-        {
-            "name": "seed",
-            "description": "A randomization seed to use for reproducibility. If not given or `null`, no seed is used and results may differ on subsequent use.",
-            "optional": true,
-            "default": null,
-            "schema": {
-                "type": [
-                    "integer",
-                    "null"
-                ]
-            }
-        }
-    ],
-    "returns": {
-        "description": "A model object that can be trained using ``ml_fit``.",
-        "schema": {
-            "type": "object",
-            "subtype": "ml-model"
-        }
-    },
-    "links": [
-        {
-            "href": "https://ieeexplore.ieee.org/document/8921180",
-            "title": "Song et al. (2019): Land Cover Classification for Satellite Images Through 1D CNN",
-            "type": "text/html",
-            "rel": "about"
-        }
-    ]
-} 
\ No newline at end of file
diff --git a/proposals/export_cube.json b/proposals/export_cube.json
deleted file mode 100644
index 8810c653..00000000
--- a/proposals/export_cube.json
+++ /dev/null
@@ -1,40 +0,0 @@
-{
-    "id": "export_cube",
-    "summary": "Export a data cube to a workspace",
-    "description": "Exports a processed data cube to a specified workspace location. This allows for storing preprocessed data (e.g., cloud-removed, gap-filled, aggregated) for later use across various machine learning models.\n\nThe workspace is recommended to be on the same server as the implementation, this ensures better performance.",
-    "categories": [
-        "Export",
-        "machine learning"
-    ],
-    "experimental": true,
-    "parameters": [
-        {
-            "name": "data",
-            "description": "The data cube to be exported. This should be a processed data cube that you want to store for later use.",
-            "schema": {
-                "type": "object",
-                "subtype": "datacube"
-            }
-        },
-        {
-            "name": "name",
-            "description": "The name to be given to the exported data cube. This will be used to identify the data when importing it later.",
-            "schema": {
-                "type": "string"
-            }
-        },
-        {
-            "name": "folder",
-            "description": "The folder path within the workspace where the data cube should be stored.",
-            "schema": {
-                "type": "string"
-            }
-        }
-    ],
-    "returns": {
-        "description": "A boolean indicating whether the export was successful.",
-        "schema": {
-            "type": "boolean"
-        }
-    }
-} 
\ No newline at end of file

From 1a506536c8a5f3b024ad9394370cf3d8b34a44f5 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Sun, 8 Feb 2026 22:24:07 +0100
Subject: [PATCH 50/52] refactor save ml model

---
 proposals/save_ml_model.json | 23 -----------------------
 1 file changed, 23 deletions(-)

diff --git a/proposals/save_ml_model.json b/proposals/save_ml_model.json
index af593363..469622e9 100644
--- a/proposals/save_ml_model.json
+++ b/proposals/save_ml_model.json
@@ -23,29 +23,6 @@
                 "type": "string"
             }
         },
-        {
-            "name": "tasks",
-            "description": "Specifies the ML tasks which the model can be used for.",
-            "schema": {
-                "type": "array",
-                "items": {
-                    "type": "string",
-                    "enum": [
-                        "regression",
-                        "classification",
-                        "object-detection",
-                        "detection",
-                        "scene-classification",
-                        "segmentation",
-                        "semantic-segmentation",
-                        "similarity-search",
-                        "generative",
-                        "image-captioning",
-                        "super-resolution"
-                    ]
-                }
-            }
-        },
         {
             "name": "options",
             "description": "Additional parameters to create the file(s).",

From 2b32cd901b9e14954676b1da661e6c2ac1a5f344 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Wed, 18 Feb 2026 15:25:43 +0100
Subject: [PATCH 51/52] ml tune and validate

---
 proposals/ml_tune_grid.json   | 146 +++++++++++++++++++++++++++
 proposals/ml_tune_random.json | 185 ++++++++++++++++++++++++++++++++++
 proposals/ml_validate.json    | 102 +++++++++++++++++++
 tests/.words                  |  20 +++-
 4 files changed, 452 insertions(+), 1 deletion(-)
 create mode 100644 proposals/ml_tune_grid.json
 create mode 100644 proposals/ml_tune_random.json
 create mode 100644 proposals/ml_validate.json

diff --git a/proposals/ml_tune_grid.json b/proposals/ml_tune_grid.json
new file mode 100644
index 00000000..b19ad755
--- /dev/null
+++ b/proposals/ml_tune_grid.json
@@ -0,0 +1,146 @@
+{
+    "id": "ml_tune_grid",
+    "summary": "Grid search hyperparameter tuning",
+    "description": "Performs exhaustive grid search over specified hyperparameter combinations to find optimal model settings. Each combination in the parameter grid is evaluated using cross-validation or a hold-out validation set, and the best trained model based on the specified scoring metrics is returned.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "model",
+            "description": "An untrained machine learning model to be trained and evaluated for each hyperparameter combination.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "training_set",
+            "description": "The training set for the model, provided as a vector data cube or a reference. This set contains both the independent variables and the dependent variable that the model analyzes to learn patterns and relationships within the data.",
+            "schema": [
+                {
+                    "type": "object",
+                    "subtype": "datacube",
+                    "dimensions": [
+                        {
+                            "type": "geometry"
+                        },
+                        {
+                            "type": "bands"
+                        }
+                    ]
+                },
+                {
+                    "type": "object",
+                    "subtype": "datacube",
+                    "dimensions": [
+                        {
+                            "type": "geometry"
+                        },
+                        {
+                            "type": "other"
+                        }
+                    ]
+                },
+                {
+                    "type": "string"
+                }
+            ]
+        },
+        {
+            "name": "parameters",
+            "description": "Hyperparameter grid defining the search space. Each key is a hyperparameter name, and each value is an array of candidate values to try. All combinations will be evaluated.",
+            "schema": {
+                "type": "object",
+                "additionalProperties": {
+                    "type": "array",
+                    "items": {}
+                }
+            }
+        },
+        {
+            "name": "target",
+            "description": "The name of the variable in the training set that serves as the target or ground truth for model training. This may refer to class labels, continuous values, segmentation masks, or any reference variable depending on the model task. Set to `null` for unsupervised or self-supervised tasks.",
+            "optional": true,
+            "default": "label",
+            "schema": [
+                {
+                    "type": "string"
+                },
+                {
+                    "type": "null"
+                }
+            ]
+        },
+        {
+            "name": "scoring",
+            "description": "One or more metrics used to evaluate and compare model performance. The available metrics depend on the model task and the back-end implementation. Examples include 'accuracy', 'f1', 'kappa', and 'auc' for classification; 'rmse', 'mae', and 'r2' for regression; 'iou' and 'dice' for segmentation; or any custom metric supported by the back-end. If set to `null`, the back-end selects an appropriate default metric based on the model task. When multiple metrics are provided, the first metric in the list is used to select the best model.",
+            "optional": true,
+            "default": null,
+            "schema": [
+                {
+                    "type": "string"
+                },
+                {
+                    "type": "array",
+                    "items": {
+                        "type": "string"
+                    },
+                    "minItems": 1
+                },
+                {
+                    "type": "null"
+                }
+            ]
+        },
+        {
+            "name": "cv",
+            "description": "Number of cross-validation folds. If 0 or 1, uses a single train/validation split instead of cross-validation.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "validation_split",
+            "description": "Fraction of training data used for validation when cv <= 1. Ignored when using cross-validation.",
+            "optional": true,
+            "default": 0.2,
+            "schema": {
+                "type": "number",
+                "minimum": 0,
+                "exclusiveMaximum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "Random seed for reproducibility of data splits and model training.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A trained model with the best hyperparameter combination found during the grid search. The model metadata includes the tuning results such as the best hyperparameters and evaluation scores.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://www.jmlr.org/papers/v13/bergstra12a.html",
+            "title": "Bergstra & Bengio (2012) - Random Search for Hyper-Parameter Optimization",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
diff --git a/proposals/ml_tune_random.json b/proposals/ml_tune_random.json
new file mode 100644
index 00000000..9d7274c2
--- /dev/null
+++ b/proposals/ml_tune_random.json
@@ -0,0 +1,185 @@
+{
+    "id": "ml_tune_random",
+    "summary": "Random search hyperparameter tuning",
+    "description": "Performs random search over hyperparameter distributions to find optimal model settings. Unlike grid search, random search samples a fixed number of parameter combinations from specified distributions, which can be more efficient for high-dimensional parameter spaces. Returns the best trained model based on the specified scoring metrics.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "model",
+            "description": "An untrained machine learning model to be trained and evaluated for each sampled hyperparameter combination.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "training_set",
+            "description": "The training set for the model, provided as a vector data cube or a reference. This set contains both the independent variables and the dependent variable that the model analyzes to learn patterns and relationships within the data.",
+            "schema": [
+                {
+                    "type": "object",
+                    "subtype": "datacube",
+                    "dimensions": [
+                        {
+                            "type": "geometry"
+                        },
+                        {
+                            "type": "bands"
+                        }
+                    ]
+                },
+                {
+                    "type": "object",
+                    "subtype": "datacube",
+                    "dimensions": [
+                        {
+                            "type": "geometry"
+                        },
+                        {
+                            "type": "other"
+                        }
+                    ]
+                },
+                {
+                    "type": "string"
+                }
+            ]
+        },
+        {
+            "name": "parameters",
+            "description": "Hyperparameter distributions defining the search space. Each key is a hyperparameter name. Values can be arrays (uniform sampling), or objects specifying distributions with 'type' ('uniform', 'log_uniform', 'int_uniform', 'choice') and parameters ('min', 'max', 'values').",
+            "schema": {
+                "type": "object",
+                "additionalProperties": {
+                    "anyOf": [
+                        {
+                            "type": "array",
+                            "description": "List of discrete values to sample from uniformly.",
+                            "items": {}
+                        },
+                        {
+                            "type": "object",
+                            "description": "Distribution specification.",
+                            "properties": {
+                                "type": {
+                                    "type": "string",
+                                    "enum": [
+                                        "uniform",
+                                        "log_uniform",
+                                        "int_uniform",
+                                        "choice"
+                                    ]
+                                },
+                                "min": {
+                                    "type": "number"
+                                },
+                                "max": {
+                                    "type": "number"
+                                },
+                                "values": {
+                                    "type": "array"
+                                }
+                            }
+                        }
+                    ]
+                }
+            }
+        },
+        {
+            "name": "n_iter",
+            "description": "Number of random parameter combinations to sample and evaluate.",
+            "optional": true,
+            "default": 10,
+            "schema": {
+                "type": "integer",
+                "minimum": 1
+            }
+        },
+        {
+            "name": "target",
+            "description": "The name of the variable in the training set that serves as the target or ground truth for model training. This may refer to class labels, continuous values, segmentation masks, or any reference variable depending on the model task. Set to `null` for unsupervised or self-supervised tasks.",
+            "optional": true,
+            "default": "label",
+            "schema": [
+                {
+                    "type": "string"
+                },
+                {
+                    "type": "null"
+                }
+            ]
+        },
+        {
+            "name": "scoring",
+            "description": "One or more metrics used to evaluate and compare model performance. The available metrics depend on the model task and the back-end implementation. Examples include 'accuracy', 'f1', 'kappa', and 'auc' for classification; 'rmse', 'mae', and 'r2' for regression; 'iou' and 'dice' for segmentation; or any custom metric supported by the back-end. If set to `null`, the back-end selects an appropriate default metric based on the model task. When multiple metrics are provided, the first metric in the list is used to select the best model.",
+            "optional": true,
+            "default": null,
+            "schema": [
+                {
+                    "type": "string"
+                },
+                {
+                    "type": "array",
+                    "items": {
+                        "type": "string"
+                    },
+                    "minItems": 1
+                },
+                {
+                    "type": "null"
+                }
+            ]
+        },
+        {
+            "name": "cv",
+            "description": "Number of cross-validation folds. If 0 or 1, uses a single train/validation split instead of cross-validation.",
+            "optional": true,
+            "default": 5,
+            "schema": {
+                "type": "integer",
+                "minimum": 0
+            }
+        },
+        {
+            "name": "validation_split",
+            "description": "Fraction of training data used for validation when cv <= 1. Ignored when using cross-validation.",
+            "optional": true,
+            "default": 0.2,
+            "schema": {
+                "type": "number",
+                "minimum": 0,
+                "exclusiveMaximum": 1
+            }
+        },
+        {
+            "name": "seed",
+            "description": "Random seed for reproducibility of parameter sampling, data splits, and model training.",
+            "optional": true,
+            "default": null,
+            "schema": {
+                "type": [
+                    "integer",
+                    "null"
+                ]
+            }
+        }
+    ],
+    "returns": {
+        "description": "A trained model with the best hyperparameter combination found during the random search. The model metadata includes the tuning results such as the best hyperparameters and evaluation scores.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://www.jmlr.org/papers/v13/bergstra12a.html",
+            "title": "Bergstra & Bengio (2012) - Random Search for Hyper-Parameter Optimization",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
diff --git a/proposals/ml_validate.json b/proposals/ml_validate.json
new file mode 100644
index 00000000..54aa4eb3
--- /dev/null
+++ b/proposals/ml_validate.json
@@ -0,0 +1,102 @@
+{
+    "id": "ml_validate",
+    "summary": "Validate a trained ML model",
+    "description": "Evaluates a trained machine learning model on a validation set and computes performance metrics. The model must have been previously trained using a process such as ``ml_fit()``, ``ml_tune_grid()``, or ``ml_tune_random()``.",
+    "categories": [
+        "machine learning"
+    ],
+    "experimental": true,
+    "parameters": [
+        {
+            "name": "model",
+            "description": "A trained machine learning model to evaluate.",
+            "schema": {
+                "type": "object",
+                "subtype": "ml-model"
+            }
+        },
+        {
+            "name": "validation_set",
+            "description": "The validation set used to evaluate the model, provided as a vector data cube or a reference. This set must contain the same independent variables the model was trained on, along with the expected target variable for comparison.",
+            "schema": [
+                {
+                    "type": "object",
+                    "subtype": "datacube",
+                    "dimensions": [
+                        {
+                            "type": "geometry"
+                        },
+                        {
+                            "type": "bands"
+                        }
+                    ]
+                },
+                {
+                    "type": "object",
+                    "subtype": "datacube",
+                    "dimensions": [
+                        {
+                            "type": "geometry"
+                        },
+                        {
+                            "type": "other"
+                        }
+                    ]
+                },
+                {
+                    "type": "string"
+                }
+            ]
+        },
+        {
+            "name": "target",
+            "description": "The name of the variable in the validation set that serves as the ground truth for evaluation. This may refer to class labels, continuous values, segmentation masks, or any reference variable depending on the model task. Set to `null` for unsupervised or self-supervised tasks.",
+            "optional": true,
+            "default": "label",
+            "schema": [
+                {
+                    "type": "string"
+                },
+                {
+                    "type": "null"
+                }
+            ]
+        },
+        {
+            "name": "scoring",
+            "description": "One or more metrics to compute for evaluating model performance. The available metrics depend on the model task and the back-end implementation. Examples include 'accuracy', 'f1', 'kappa', and 'auc' for classification; 'rmse', 'mae', and 'r2' for regression; 'iou' and 'dice' for segmentation; or any custom metric supported by the back-end. If set to `null`, the back-end selects an appropriate default metric based on the model task.",
+            "optional": true,
+            "default": null,
+            "schema": [
+                {
+                    "type": "string"
+                },
+                {
+                    "type": "array",
+                    "items": {
+                        "type": "string"
+                    },
+                    "minItems": 1
+                },
+                {
+                    "type": "null"
+                }
+            ]
+        }
+    ],
+    "returns": {
+        "description": "The trained model enriched with validation scores in its metadata. The model can be directly used for prediction with ``ml_predict()``.",
+        "schema": {
+            "type": "object",
+            "subtype": "ml-model"
+        }
+    },
+    "links": [
+        {
+            "href": "https://doi.org/10.1214/09-SS054",
+            "title": "Arlot & Celisse (2010) - A Survey of Cross-Validation Procedures for Model Selection",
+            "type": "text/html",
+            "rel": "about"
+        }
+    ]
+}
diff --git a/tests/.words b/tests/.words
index 25ba9e5e..d5ac5038 100644
--- a/tests/.words
+++ b/tests/.words
@@ -83,4 +83,22 @@ Khanna
 Awad
 CatBoost
 epsilon-SVR
-Terratorch
\ No newline at end of file
+Terratorch
+hyperparameter
+hyperparameters
+Hyperparameter
+f1
+auc
+rmse
+mse
+mae
+r2
+cv
+Bergstra
+Bengio
+Arlot
+Celisse
+log_uniform
+int_uniform
+iou
+dice
\ No newline at end of file

From 5d41223e99e44dad3796704fda4c44efa372afc3 Mon Sep 17 00:00:00 2001
From: Brian Pondi <pondibrian@gmail.com>
Date: Fri, 20 Feb 2026 09:24:11 +0100
Subject: [PATCH 52/52] save ml variable renaming

---
 proposals/save_ml_model.json | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/proposals/save_ml_model.json b/proposals/save_ml_model.json
index 469622e9..ab5820ba 100644
--- a/proposals/save_ml_model.json
+++ b/proposals/save_ml_model.json
@@ -9,8 +9,8 @@
     "experimental": true,
     "parameters": [
         {
-            "name": "data",
-            "description": "The data to store as a machine learning model.",
+            "name": "model",
+            "description": "The machine learning model to store.",
             "schema": {
                 "type": "object",
                 "subtype": "ml-model"