This is a python3
implementation of the neural network discussed in Michael Nielsen's book
The neural net is used as a classifier for handwritten digits .
First you need pipenv to be installed globally, for that, run pip install pipenv
Then for installing the relevant packages for this repository, run pipenv install
Which will create a virtual environment dedicated to this repository with all packages installed.
Now we have to spawn a shell inside that virtual environment to start working, to do so run pipenv shell
Simply run py train.py
Which will simulate training epochs and print progress to the console.
If you have read nielson's book then the code should be self explanatory, but If you didn't then i recommend you do, because there's a lot of information to explain in a README .
A pull request is welcome any time .