All Projects → NVIDIA → transformer-ls

NVIDIA / transformer-ls

Licence: MIT license
Official PyTorch Implementation of Long-Short Transformer (NeurIPS 2021).

Programming Languages

python
139335 projects - #7 most used programming language
shell
77523 projects
Dockerfile
14818 projects

Projects that are alternatives of or similar to transformer-ls

keras-vision-transformer
The Tensorflow, Keras implementation of Swin-Transformer and Swin-UNET
Stars: ✭ 91 (-54.73%)
Mutual labels:  transformer, vision-transformer
image-classification
A collection of SOTA Image Classification Models in PyTorch
Stars: ✭ 70 (-65.17%)
Mutual labels:  transformer, vision-transformer
TransMorph Transformer for Medical Image Registration
TransMorph: Transformer for Unsupervised Medical Image Registration (PyTorch)
Stars: ✭ 130 (-35.32%)
Mutual labels:  transformer, vision-transformer
YOLOS
You Only Look at One Sequence (NeurIPS 2021)
Stars: ✭ 612 (+204.48%)
Mutual labels:  transformer, vision-transformer
SwinIR
SwinIR: Image Restoration Using Swin Transformer (official repository)
Stars: ✭ 1,260 (+526.87%)
Mutual labels:  transformer, vision-transformer
semantic-segmentation
SOTA Semantic Segmentation Models in PyTorch
Stars: ✭ 464 (+130.85%)
Mutual labels:  transformer, vision-transformer
towhee
Towhee is a framework that is dedicated to making neural data processing pipelines simple and fast.
Stars: ✭ 821 (+308.46%)
Mutual labels:  transformer, vision-transformer
SReT
Official PyTorch implementation of our ECCV 2022 paper "Sliced Recursive Transformer"
Stars: ✭ 51 (-74.63%)
Mutual labels:  vision-transformer, efficient-transformers
Restormer
[CVPR 2022--Oral] Restormer: Efficient Transformer for High-Resolution Image Restoration. SOTA for motion deblurring, image deraining, denoising (Gaussian/real data), and defocus deblurring.
Stars: ✭ 586 (+191.54%)
Mutual labels:  transformer, efficient-transformers
LaTeX-OCR
pix2tex: Using a ViT to convert images of equations into LaTeX code.
Stars: ✭ 1,566 (+679.1%)
Mutual labels:  transformer, vision-transformer
visualization
a collection of visualization function
Stars: ✭ 189 (-5.97%)
Mutual labels:  transformer, vision-transformer
VT-UNet
[MICCAI2022] This is an official PyTorch implementation for A Robust Volumetric Transformer for Accurate 3D Tumor Segmentation
Stars: ✭ 151 (-24.88%)
Mutual labels:  transformer, vision-transformer
Ghostnet
CV backbones including GhostNet, TinyNet and TNT, developed by Huawei Noah's Ark Lab.
Stars: ✭ 1,744 (+767.66%)
Mutual labels:  transformer, vision-transformer
libai
LiBai(李白): A Toolbox for Large-Scale Distributed Parallel Training
Stars: ✭ 284 (+41.29%)
Mutual labels:  transformer, vision-transformer
ViTs-vs-CNNs
[NeurIPS 2021]: Are Transformers More Robust Than CNNs? (Pytorch implementation & checkpoints)
Stars: ✭ 145 (-27.86%)
Mutual labels:  transformer
kaggle-champs
Code for the CHAMPS Predicting Molecular Properties Kaggle competition
Stars: ✭ 49 (-75.62%)
Mutual labels:  transformer
R-MeN
Transformer-based Memory Networks for Knowledge Graph Embeddings (ACL 2020) (Pytorch and Tensorflow)
Stars: ✭ 74 (-63.18%)
Mutual labels:  transformer
TokenLabeling
Pytorch implementation of "All Tokens Matter: Token Labeling for Training Better Vision Transformers"
Stars: ✭ 385 (+91.54%)
Mutual labels:  transformer
cape
Continuous Augmented Positional Embeddings (CAPE) implementation for PyTorch
Stars: ✭ 29 (-85.57%)
Mutual labels:  transformer
Representation-Learning-for-Information-Extraction
Pytorch implementation of Paper by Google Research - Representation Learning for Information Extraction from Form-like Documents.
Stars: ✭ 82 (-59.2%)
Mutual labels:  transformer

Long-Short Transformer (Transformer-LS)

This repository hosts the code and models for the paper:

Long-Short Transformer: Efficient Transformers for Language and Vision

Updates

Architecture

plot Long-short Transformer substitutes the full self attention of the original Transformer models with an efficient attention that considers both long-range and short-term correlations. Each query attends to tokens from the segment-wise sliding window to capture short-term correlations, and the dynamically projected features to capture long-range correlations. To align the norms of the original and projected feature vectors and improve the efficacy of the aggregation, we normalize the original and project feature vectors with two sets of Layer Normalizations.

Tasks

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].