deit

A treasure chest for visual classification and recognition powered by PaddlePaddle

PASSL包含 SimCLR，MoCo v1/v2，BYOL，CLIP，PixPro，simsiam, SwAV, BEiT，MAE 等图像自监督算法以及 Vision Transformer，DEiT，Swin Transformer，CvT，T2T-ViT，MLP-Mixer，XCiT，ConvNeXt，PVTv2 等基础视觉算法

Paddle Large Scale Classification Tools，supports ArcFace, CosFace, PartialFC, Data Parallel + Model Parallel. Model includes ResNet, ViT, Swin, DeiT, CaiT, FaceViT, MoCo, MAE, ConvMAE, CAE.

A PaddlePaddle version image model zoo.

(Unofficial) PyTorch implementation of Training Vision Transformers for Image Retrieval(El-Nouby, Alaaeldin, et al. 2021).

[CVPR 2024] Code for our Paper "DeiT-LT: Distillation Strikes Back for Vision Transformer training on Long-Tailed Datasets"

Free & Open Source. Precise and flexible garment measurements from images - no tape measures, no delays, just fashion - forward automation.

2D Human Pose estimation using transformers. Implementation in Pytorch

Pytorch implementation of some vision transformers, trained on CIFAR-10.

[CVPR'24] Once for Both: Single Stage of Importance and Sparsity Search for Vision Transformer Compression

Image Classification Tutorial: ConvNext--> 98.8% on CIFAR10 + 92.4% on CIFAR100; ResNet18 -- 95.6% on CIFAR10 + 79.1% on CIFAR100

This is a warehouse for DeiT-pytorch-model, can be used to train your image dataset

Project on exploring how different vision models “see” and analyze the images. We compare ConvNeXt, DeiT, and MLP-Mixer using Grad-CAM, attention maps, and saliency, with tools to run inference, analyze results, and visualize model focus regions.

Multi-Label Retinal Disease Classification using Vision Transformers (Swin, DeiT, ViT) vs ResNet-50 | 81.1% F1 Macro | PyTorch Lightning

VisionTransformer for Tensorflow2

Final assignment in the NLP course at the Technion (IEM097215). In this assignment we propose a novel architecture to handle both Text-to-Image translation and Image-to-Text translation tasks on paired data, using a unified architecture of transformers and CNNs and enforcing cycle consistency.

Swin Transformer + DETR, applied distillation queries & hard-level distillation

Benchmarking CNN and Vision Transformer architectures for geospatial land-use classification. Features ResNet-18 (SE), DeiT-3, and Normalization-Free Transformers (DyT) with Grad-CAM and Attention Map analysis.

Official PyTorch implementation of “PLuG: Pairwise Logit Gating for Expressive Attention Modulation in Vision Transformers” (IEEE Access 2026).

"Neural Computing and Applications" Published Paper (2023)