berniwal/swin-transformer-pytorch
PyTorch implementation of Swin Transformer, a hierarchical vision transformer for computer vision tasks.
Not currently ranked — collecting fresh signals.
star history
This repository provides a PyTorch implementation of the Swin Transformer architecture, a vision transformer designed for computer vision tasks. The model uses a shifted windowing scheme for self-attention to achieve linear computational complexity with respect to image size. It serves as a general-purpose backbone for tasks including image classification, object detection, and semantic segmentation, achieving strong performance on benchmarks such as ImageNet-1K, COCO, and ADE20K.
Frequently asked
- What is berniwal/swin-transformer-pytorch?
- PyTorch implementation of Swin Transformer, a hierarchical vision transformer for computer vision tasks.
- Is swin-transformer-pytorch open source?
- Yes — berniwal/swin-transformer-pytorch is open source, released under the MIT license.
- What language is swin-transformer-pytorch written in?
- berniwal/swin-transformer-pytorch is primarily written in Python.
- How popular is swin-transformer-pytorch?
- berniwal/swin-transformer-pytorch has 862 stars on GitHub.
- Where can I find swin-transformer-pytorch?
- berniwal/swin-transformer-pytorch is on GitHub at https://github.com/berniwal/swin-transformer-pytorch.