swikargautam Goto Github PK

call-break

convonet

A framework for implementing convolutional neural networks and fully connected neural network.

deepdensenetwork

Fully connected neural network with Adam optimizer, L2 regularization, Batch normalization, and Dropout using only numpy

gan

generative adversarial network trained on mnist images

neat

Python implementation of NEAT(NeuroEvolution of Augmenting Topologies)

rl_misc

Implementation of some RL algorithms from the book "Reinforcement Learning: An Introduction" by Sutton and Barto

transformer

Implementation of vanilla transformer model using numpy.

vae

Variational Autoencoder (VAE) trained on MNIST