Deep learning using tensorflow
#Tensorflow Projects A repo of everything deep and neurally related. Implementations and ideas are largely based on papers from arxiv and implementations, tutorials from the internet.
##Folders
##Results
Generative Adversarial Networks - Tensorflow implementation of DCGAN on Flowers dataset and celebA dataset. Used carpedm20 and Newmu's DCGAN code for reference. The sample outputs below are the generated images - Results in increading order of epochs from left to right.
CelebA
But the celebA dataset used here had faces that are well aligned - Can we do better with unaligned images? Well, the results are as below.
Flowers
Face Inpainting - Tried Context Encoder on Labeled Faces in Wild dataset and these are the results
Deep dreams
=== Checkout my neuralnetworks.thought-experiments for more in depth notes on Deep learning models and concepts.