Installation Issue - no valid installation of tensorflow

[lynseymccoll]

Hi,

I'm having problems installing tensorflow.

I can add the following code:

install_tensorflow(envname = "C:\Users\LynseyMcColl\AppData\Local\r-tensorflow", method = "conda", python_version = "3.11", version = "2.18")

Which says the installation is successful but when I then try it out I get the following error message:
Error: Valid installation of TensorFlow not found.

I have set an envname above as I am working on one drive and so have pointed it to the desktop rather than an online environment which I've read can cause problems.

I've put the messages I get to the commands requested below.

Many thanks,

Lynsey

reticulate::py_config()

libpython:      C:/Users/LynseyMcColl/AppData/Local/Programs/Python/Python311/python311.dll
pythonhome:     C:/Users/LynseyMcColl/AppData/Local/Programs/Python/Python311
version:        3.11.0 (main, Oct 24 2022, 18:26:48) [MSC v.1933 64 bit (AMD64)]
Architecture:   64bit
numpy:           [NOT FOUND]

NOTE: Python version was forced by use_python() function

tensorflow::tf_config()

Valid installation of TensorFlow not found.

Python environments searched for 'tensorflow' package:
 C:\Users\LynseyMcColl\AppData\Local\Programs\Python\Python311\python.exe

Python exception encountered:
 Traceback (most recent call last):
  File "C:\Users\LynseyMcColl\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 122, in _find_and_load_hook
    return _run_hook(name, _hook)
           ^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\LynseyMcColl\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 96, in _run_hook
    module = hook()
             ^^^^^^
  File "C:\Users\LynseyMcColl\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 120, in _hook
    return _find_and_load(name, import_)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ModuleNotFoundError: No module named 'tensorflow'


You can install TensorFlow using the install_tensorflow() function.
 
Warning message:
ModuleNotFoundError: No module named 'tensorflow'
Run `reticulate::py_last_error()` for details.
Restart the R session and load the tensorflow R package before reticulate has initialized Python, or ensure reticulate initialized a Python installation where the tensorflow module is installed.

reticulate::import("tensorflow")

Error in py_module_import(module, convert = convert) : 
  ModuleNotFoundError: No module named 'tensorflow'
Run `reticulate::py_last_error()` for details.

reticulate::py_last_error()

── Python Exception Message ──────────────────────────────────────────────────────────────────────────────────
Traceback (most recent call last):
  File "C:\Users\LynseyMcColl\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 122, in _find_and_load_hook
    return _run_hook(name, _hook)
           ^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\LynseyMcColl\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 96, in _run_hook
    module = hook()
             ^^^^^^
  File "C:\Users\LynseyMcColl\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 120, in _hook
    return _find_and_load(name, import_)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ModuleNotFoundError: No module named 'tensorflow'

── R Traceback ───────────────────────────────────────────────────────────────────────────────────────────────
    ▆
 1. └─reticulate::import("tensorflow")
 2.   └─reticulate:::py_module_import(module, convert = convert)
See `reticulate::py_last_error()$r_trace$full_call` for more details.

sessionInfo()

R version 4.4.3 (2025-02-28 ucrt)
Platform: x86_64-w64-mingw32/x64
Running under: Windows 10 x64 (build 19045)

Matrix products: default


locale:
[1] LC_COLLATE=English_United Kingdom.utf8  LC_CTYPE=English_United Kingdom.utf8   
[3] LC_MONETARY=English_United Kingdom.utf8 LC_NUMERIC=C                           
[5] LC_TIME=English_United Kingdom.utf8    

time zone: Europe/London
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

loaded via a namespace (and not attached):
 [1] base64enc_0.1-3        Matrix_1.7-2           lattice_0.22-6         reticulate_1.42.0     
 [5] magrittr_2.0.3         glue_1.8.0             png_0.1-8              lifecycle_1.0.4       
 [9] cli_3.6.4              tensorflow_2.16.0.9000 vctrs_0.6.5            grid_4.4.3            
[13] tfruns_1.5.3           compiler_4.4.3         rstudioapi_0.17.1      tools_4.4.3           
[17] whisker_0.4.1          pillar_1.10.1          Rcpp_1.0.14            rlang_1.1.5           
[21] jsonlite_2.0.0   

[t-kalinowski]

What happens if you run the following:

In a fresh R session:

reticulate:::rm_all_reticulate_state()
remotes::install_github("rstudio/tensorflow")

And then, in another fresh R session:

Sys.setenv("RETICULATE_PYTHON" = "managed")
library(tensorflow)
tf$constant("Hello World")

[lynseymccoll]

Thanks for getting back to me so quickly.

Here is what I get:

> reticulate:::rm_all_reticulate_state()
> remotes::install_github("rstudio/tensorflow")
Skipping install of 'tensorflow' from a github remote, the SHA1 (e48867f4) has not changed since last install.
  Use `force = TRUE` to force installation

Restarting R session...


> tf$constant("Hello World")
Downloading cpython-3.11.12-windows-x86_64-none (24.3MiB)
 Downloaded cpython-3.11.12-windows-x86_64-none
Downloading pygments (1.2MiB)
Downloading h5py (2.8MiB)
Downloading tensorflow-io-gcs-filesystem (1.4MiB)
Downloading setuptools (1.2MiB)
Downloading libclang (25.2MiB)
Downloading numpy (15.1MiB)
Downloading tensorflow (358.5MiB)
Downloading grpcio (4.1MiB)
Downloading keras (1.3MiB)
Downloading tensorboard (5.2MiB)
 Downloaded tensorflow-io-gcs-filesystem
 Downloaded h5py
 Downloaded grpcio
 Downloaded pygments
 Downloaded setuptools
 Downloaded tensorboard
 Downloaded keras
 Downloaded numpy
 Downloaded libclang
 Downloaded tensorflow
Installed 38 packages in 24.84s

Error: Valid installation of TensorFlow not found.  Python environments searched for 'tensorflow' package:  C:\Users\LynseyMcColl\AppData\Local\R\cache\R\reticulate\uv\cache\archive-v0\yr1j5_ZBEj9Mt4o17V7_J\Scripts\python.exe  Python exception encountered:  Traceback (most recent call last):   File "C:\Users\LYNSEY~1\AppData\Local\R\cache\R\RETICU~1\uv\cache\ARCHIV~1\YR1J5_~1\Lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 73, in <module>     from tensorflow.python._pywrap_tensorflow_internal import *   File "C:\Users\LynseyMcColl\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 122, in _find_and_load_hook     return _run_hook(name, _hook)            ^^^^^^^^^^^^^^^^^^^^^^   File "C:\Users\LynseyMcColl\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 96, in _run_hook     module = hook()              ^^^^^^   File "C:\Users\LynseyMcColl\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 120, i
--

I'm sorry if I've done something strange with the installation.

Lynsey

[t-kalinowski]

Thanks, I can reproduce the same error.
There is something wrong with the recent TensorFlow builds for Windows that will take some investigation.

For now you can workaround by adding a call to py_require(exclude_newer = ) before python has initialized:

Sys.setenv("RETICULATE_PYTHON" = "managed")
reticulate::py_require(exclude_newer = "2025-03-11")

library(tensorflow)
tf$constant("Hello World")

[lynseymccoll]

Yes I think this works now.

I get:

'''

tf$constant("Hello World")
2025-04-10 14:06:33.321594: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable TF_ENABLE_ONEDNN_OPTS=0.
2025-04-10 14:06:35.051642: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable TF_ENABLE_ONEDNN_OPTS=0.
2025-04-10 14:06:38.059908: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
tf.Tensor(b'Hello World', shape=(), dtype=string)

Thank you for reviewing and I hope the issue is easily fixed.

Lynsey

[t-kalinowski]

This error seems to only happen in the RStudio IDE. I can't reproduce it in Rterm.exe

minimal reprex:

tensorflow::install_tensorflow()

library(reticulate)
tf <- import("tensorflow")
tf$constant("Hello World")

Produces:

> library(reticulate)
> tf <- import("tensorflow")
Error in py_module_import(module, convert = convert) : 
  ImportError: Traceback (most recent call last):
  File "C:\Users\kalin\DOCUME~1\VIRTUA~1\R-TENS~1\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 70, in <module>
<...truncated...>ppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 96, in _run_hook
    module = hook()
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 120, in _hook
    return _find_and_load(name, import_)
ImportError: DLL load failed while importing _pywrap_tensorflow_internal: A dynamic link library (DLL) initialization routine failed.


Failed to load the native TensorFlow runtime.
See https://www.tensorflow.org/install/errors for some common causes and solutions.
If you need help, create an issue at https://github.com/tensorflow/tensorflow/issues and include the entire stack trace above this error message.
Run `reticulate::py_last_error()` for details.

> reticulate::py_last_error()

── Python Exception Message ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Traceback (most recent call last):
  File "C:\Users\kalin\DOCUME~1\VIRTUA~1\R-TENS~1\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 70, in <module>
    from tensorflow.python._pywrap_tensorflow_internal import *
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 122, in _find_and_load_hook
    return _run_hook(name, _hook)
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 96, in _run_hook
    module = hook()
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 120, in _hook
    return _find_and_load(name, import_)
ImportError: DLL load failed while importing _pywrap_tensorflow_internal: A dynamic link library (DLL) initialization routine failed.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 122, in _find_and_load_hook
    return _run_hook(name, _hook)
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 96, in _run_hook
    module = hook()
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 120, in _hook
    return _find_and_load(name, import_)
  File "C:\Users\kalin\DOCUME~1\VIRTUA~1\R-TENS~1\lib\site-packages\tensorflow\__init__.py", line 40, in <module>
    from tensorflow.python import pywrap_tensorflow as _pywrap_tensorflow  # pylint: disable=unused-import
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 122, in _find_and_load_hook
    return _run_hook(name, _hook)
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 96, in _run_hook
    module = hook()
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 120, in _hook
    return _find_and_load(name, import_)
  File "C:\Users\kalin\DOCUME~1\VIRTUA~1\R-TENS~1\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 85, in <module>
    raise ImportError(
ImportError: Traceback (most recent call last):
  File "C:\Users\kalin\DOCUME~1\VIRTUA~1\R-TENS~1\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 70, in <module>
    from tensorflow.python._pywrap_tensorflow_internal import *
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 122, in _find_and_load_hook
    return _run_hook(name, _hook)
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 96, in _run_hook
    module = hook()
  File "C:\Users\kalin\AppData\Local\R\win-library\4.4\reticulate\python\rpytools\loader.py", line 120, in _hook
    return _find_and_load(name, import_)
ImportError: DLL load failed while importing _pywrap_tensorflow_internal: A dynamic link library (DLL) initialization routine failed.


Failed to load the native TensorFlow runtime.
See https://www.tensorflow.org/install/errors for some common causes and solutions.
If you need help, create an issue at https://github.com/tensorflow/tensorflow/issues and include the entire stack trace above this error message.

── R Traceback ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
    ▆
 1. └─reticulate::import("tensorflow")
 2.   └─reticulate:::py_module_import(module, convert = convert)
See `reticulate::py_last_error()$r_trace$full_call` for more details.

[lynseymccoll]

Do you mean the original error?

I think the work around version is allowing me to use it OK.

Thanks,

Lynsey

[t-kalinowski]

Sorry, I posted the above for others that come across this, (including future me).
Please continue to use the exclude_newer workaround.