keyu-tian

SparK

[ICLR'23 Spotlight🔥] The first successful BERT/MAE-style pretraining on any convolutional network; Pytorch impl. of "Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling"

BUAA-datastructure-project-solution

[Ranked No. 1🥇] My solution for the course project of Datastructure 2019'Spring @ BUAA (北航数据结构). Plenty of C language tricks, hacks, and optimizations are used for extreme efficiency. *Ranked 1/800* in the efficiency test.

BUAA-parallel-computing-project-solution

[Ranked No. 1🥇] My solution for the course project of Parallel Computing 2021'Spring @ BUAA (北航并行程序设计). Plenty of C++ tricks, hacks, and optimizations are used for extreme efficiency. Ranked *1/100* in the efficiency test.

EVA-CLIP

EVA Series: Visual Representation Fantasies from BAAI

keyu-tian

flash-attention

Fast and memory-efficient exact attention

mar

PyTorch implementation of MAR+DiffLoss https://arxiv.org/abs/2406.11838

alphatensor

biocatalysis-model

RXN for biochemical reactions

CogVideo

text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

ConvNeXt

Code release for ConvNeXt model

DAB-DETR

[ICLR 2022] DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETR

DALLE2-pytorch

Implementation of DALL-E 2, OpenAI's updated text-to-image synthesis neural network, in Pytorch

ESAN

Equivariant Subgraph Aggregation Networks (ICLR 2022 Spotlight)

GNNAK

Source code for From Stars to Subgraphs (ICLR 2022)

ibot

iBOT: Image BERT Pre-Training with Online Tokenizer

image-gpt

latent-diffusion

High-Resolution Image Synthesis with Latent Diffusion Models

mae

PyTorch implementation of MAE https//arxiv.org/abs/2111.06377

muse

Implementation of Muse: Text-to-Image Generation via Masked Generative Transformers, in Pytorch

Open-MAGVIT2

Open-MAGVIT2: Democratizing Autoregressive Visual Generation

pdf-cut-white

自动裁剪PDF图表中的白边 / Cut white bound in PDF figures automatically.

PixArt-alpha

Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis

rq-vae

The official implementation of Autoregressive Image Generation using Residual Quantization (CVPR '22)

segment-anything-2

The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

stable-diffusion

A latent text-to-image diffusion model

stable-diffusion-2

SD 2.0, High-Resolution Image Synthesis with Latent Diffusion Models

stylegan-t

[ICML'23] StyleGAN-T: Unlocking the Power of GANs for Fast Large-Scale Text-to-Image Synthesis

stylegan-xl

[SIGGRAPH'22] StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasets

vqgan

Taming Transformers for High-Resolution Image Synthesis