Skip to content
Implementation of neural network in Python using Numpy and Autograd.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.gitignore
LICENSE
README.md
cifar.py
mnist.py
neuralnetwork.py
utils.py

README.md

Neural Network

Author: Jakub Cisło
http://cislo.net.pl
jakub@cislo.net.pl

Implementation of neural network in Python using Numpy and Autograd.
Method: transformation matrix and stochastic gradient descent
Required: python2, numpy, matplotlib, autograd, MNIST and CIFAR-10 datasets

##Usage: python -i mnist.py
python -i cifar.py

###Interactive commands

  • create(layers=[28*28, 100, 10], batch_size=32, dropout=0.1) - create a new model
  • learn() - start the model learning
  • save(name) - save the model to file
  • load(name) - load the model from file
  • info() - show info and statistics about the model
  • best() - show best images from each category
  • worst() - show worst images from each category
You can’t perform that action at this time.