diff --git a/README.md b/README.md index e7be31f09..25880d3e2 100644 --- a/README.md +++ b/README.md @@ -3,15 +3,15 @@ # Overview -we will use PyTorch to build and train a deep learning model to classify images to 29 classes (26 ASL alphabet, space, Del, and nothing) which can be used later to help Deaf peope communicate with other and maybe with computers as well. We will use a pre-trained mobile network and we will define our classifier and connect it to network, then train this classifier along with some of the last blocks of the network on our dataset. +We will use PyTorch to build and train a deep learning model to classify images to 29 classes (26 ASL alphabet, space, Del, and nothing) which can be used later to help deaf people communicate with other and maybe with computers as well. We will use a pre-trained mobile network and we will define our classifier and connect it to network, then train this classifier along with some of the last blocks of the network on our dataset. In this notebook we will: 1- Obtain dataset from Kaggle. -2- explore data and define transformers to preprocess images before training. -3- define our classifier to have an output layer of 29 outputs. -4- train the last blocks of the network along with the classifier we defined. -5- test the model we trained. +2- Explore data and define transformers to preprocess images before training. +3- Define our classifier to have an output layer of 29 outputs. +4- Train the last blocks of the network along with the classifier we defined. +5- Test the model we trained. This notebook uses python 3.6 + GPU environment which allow us to do the whole process and train complex model in the same place which is a notebook in Watson studio. Learn more about available environments Watson Studio environments.