# zhanghang1989/PyTorch-Encoding

A CV toolkit for my papers.

Repository: https://github.com/zhanghang1989/PyTorch-Encoding
Canonical: https://ross.abutalabs.com/products/pytorch-encoding
Homepage: https://hangzhang.org/PyTorch-Encoding/
Language: Python
License: MIT
License Family: permissive
Topics: deep-learning, batchnorm, synchronized-batchnorm, encoding-layer
Last push: 2024-12-21T05:04:53+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3400, "days_push": 620, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2044, forks 448 (observed 2026-08-28T04:06:08.756952+00:00)

## What it is
A PyTorch computer vision toolkit with optimized CUDA backend, providing encoding layers, synchronized multi-GPU batch normalization, and pretrained models for semantic segmentation and image classification. It accompanies research papers such as ResNeSt, Context Encoding for Semantic Segmentation, and Deep TEN.

## Use cases
- run semantic segmentation on ADE20K or PASCAL Context
- use ResNeSt split-attention networks in PyTorch
- synchronized batch normalization across multiple GPUs
- pretrained image classification models for PyTorch
- texture encoding networks for deep texture recognition
- style transfer with MSG-Net

## When to choose
- you need the official implementations of ResNeSt or EncNet models
- you want synchronized multi-GPU batch normalization in PyTorch
- you need pretrained segmentation and classification model zoos

## When to avoid
- you need a general-purpose actively developed CV framework rather than paper-specific code
- you don't use PyTorch or CUDA GPUs
- you need Windows support or frequent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: deep-learning, machine-learning, image-processing, gpu-computing
- domain: deep-learning, computer-vision, image-processing, machine-learning
- platform: python
- tags: pytorch, semantic-segmentation, synchronized-batchnorm, encoding-layer, resnest, computer-vision-toolkit, cuda, linux, macos, gpu, docker

## Member repositories
- zhanghang1989/PyTorch-Encoding (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:08.756952+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:58:05.454617+00:00, confidence not recorded.
  - readme: https://github.com/zhanghang1989/PyTorch-Encoding (fetched 2026-08-28T04:06:08.756952+00:00, sha 79ad4261b60c)
  - homepage: https://hangzhang.org/PyTorch-Encoding/ (fetched 2026-08-29T10:38:21.280358+00:00, sha f50935b22a18)
  - site_page: https://hangzhang.org/PyTorch-Encoding/notes/compile.html (fetched 2026-08-29T10:38:21.289700+00:00, sha f583dbff3388)
- Data as of 2026-08-30T08:39:29.467469+00:00.