1. X TransformersA simple but complete full-attention transformer with a set of promising experimental features from various papers
2. Dalle PytorchImplementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch
5. Transformer In TransformerImplementation of Transformer in Transformer, pixel level attention paired with patch level attention for image classification, in Pytorch
6. Siren PytorchPytorch implementation of SIREN - Implicit Neural Representations with Periodic Activation Function
8. Stylegan2 PytorchSimplest working implementation of Stylegan2, state of the art generative adversarial network, in Pytorch. Enabling everyone to experience disentanglement
11. Perceiver PytorchImplementation of Perceiver, General Perception with Iterative Attention, in Pytorch
13. Lambda NetworksImplementation of LambdaNetworks, a new approach to image recognition that reaches SOTA with less compute
16. ConformerImplementation of the convolutional module from the Conformer paper, for use in Transformers
17. Se3 Transformer PytorchImplementation of SE3-Transformers for Equivariant Self-Attention, in Pytorch. This specific repository is geared towards integration with eventual Alphafold2 replication.
18. Mixture Of ExpertsA Pytorch implementation of Sparsely-Gated Mixture of Experts, for massively increasing the parameter count of language models
20. Contrastive LearnerA simple to use pytorch wrapper for contrastive self-supervised learning on any neural network
21. Isab PytorchAn implementation of (Induced) Set Attention Block, from the Set Transformers paper
22. Byol PytorchUsable Implementation of "Bootstrap Your Own Latent" self-supervised learning, from Deepmind, in Pytorch
23. Big SleepA simple command line tool for text to image generation, using OpenAI's CLIP and a BigGAN
24. Deep DazeSimple command line tool for text to image generation using OpenAI's CLIP and Siren (Implicit neural representation network)
26. Lightweight GanImplementation of 'lightweight' GAN, proposed in ICLR 2021, in Pytorch. High resolution image generations that can be trained within a day or two
28. Alphafold2To eventually become an unofficial Pytorch implementation / replication of Alphafold2, as details of the architecture get released
29. Vit PytorchImplementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch
30. Timesformer PytorchImplementation of TimeSformer from Facebook AI, a pure attention-based solution for video classification
32. ddpm-proteinsA denoising diffusion probabilistic model (DDPM) tailored for conditional generation of protein distograms
35. transganformerImplementation of TransGanFormer, an all-attention GAN that combines the finding from the recent GanFormer and TransGan paper
36. n-grammer-pytorchImplementation of N-Grammer, augmenting Transformers with latent n-grams, in Pytorch
37. rela-transformerImplementation of a Transformer using ReLA (Rectified Linear Attention) from https://arxiv.org/abs/2104.07012
38. geometric-vector-perceptronImplementation of Geometric Vector Perceptron, a simple circuit for 3d rotation equivariance for learning over large biomolecules, in Pytorch. Idea proposed and accepted at ICLR 2021
39. nuwa-pytorchImplementation of NÜWA, state of the art attention network for text to video synthesis, in Pytorch
40. enformer-pytorchImplementation of Enformer, Deepmind's attention network for predicting gene expression, in Pytorch
42. AoA-pytorchA Pytorch implementation of Attention on Attention module (both self and guided variants), for Visual Question Answering
44. halonet-pytorchImplementation of the 😇 Attention layer from the paper, Scaling Local Self-Attention For Parameter Efficient Visual Backbones
45. tr-rosetta-pytorchImplementation of trRosetta and trDesign for Pytorch, made into a convenient package, for protein structure prediction and design
47. token-shift-gptImplementation of Token Shift GPT - An autoregressive model that solely relies on shifting the sequence space for mixing
49. long-short-transformerImplementation of Long-Short Transformer, combining local and global inductive biases for attention over long sequences, in Pytorch