Simple Neural networks

src/one_layer_mnist/ - implementation of the 1 layer perceptron. it was trained against MNIST data set with 60k samples, checked against 10k samples with ~70% success

Books that I've read and can recommend (also free to download or/and read)

[DL-Bengio] Ian Goodfellow, Yoshua Bengio, Aaron Courville Deep Learning - pretty much state of art book, as well as very detailed references.

[Haykin-NNLM] S.Haykin Neural Networks and Learning Machines (3rd Edition) - quite classical from a famous author, has a lot of references. I think NND is better to start with neural networks

[ISLR] An Introduction to Statistical Learning with Applications in R - it's a really good book about statistical learning, doesnt require deep knowledge in linear agebra, yet very informative. Data sets are widly avaliable, examples are great.

[NND] M.Hagan H.Demuth - Neural Network Design (2nd Edition) - extremally cheap (~25$ on Amazon) but super usefull, also contains introduction into linear algebra

Prepare environemnt

1. Install python virtualenv
2. Init and activate virtual env

  # virtualenv ./.neuro -p python3
  # source ./.neuro/bin/activate

3. Install python libraries

  # pip install -r .requirements

Update environment

1. store python libraries list

  # pip freeze -r .requirements

Compile TensorFlow locally for MacOS with optimizations

My gist