kundan2510/pixelCNN
Theano reimplementation of pixelCNN architecture
Implements autoregressive image generation using masked convolutions to eliminate blind spots, with 2-bit quantized pixel prediction via four-way softmax. Built in core Theano with Keras for data loading and Lasagne optimizers, featuring efficient vertical/horizontal stack convolutions and residual connections for stable training on GPU.
166 stars. No commits in the last 6 months.
Stars
166
Forks
34
Language
Python
License
MIT
Category
Last pushed
Jul 23, 2016
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/kundan2510/pixelCNN"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
zhangqianhui/Conditional-GAN
Tensorflow implementation for Conditional Convolutional Adversarial Networks.
mit-han-lab/data-efficient-gans
[NeurIPS 2020] Differentiable Augmentation for Data-Efficient GAN Training
EugenHotaj/pytorch-generative
Easy generative modeling in PyTorch
shaohua0116/DCGAN-Tensorflow
A Tensorflow implementation of Deep Convolutional Generative Adversarial Networks trained on...
pytti-tools/pytti-notebook
Start here