# yassouali/pytorch-segmentation

:art: Semantic segmentation models, datasets and losses implemented in PyTorch.

Repository: https://github.com/yassouali/pytorch-segmentation
Canonical: https://ross.abutalabs.com/products/pytorch-segmentation
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: semantic-segmentation, computer-vision, deep-learning
Last push: 2025-02-23T19:05:28+00:00

## Health v2 (maintenance only)
Score: 26/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 8, release rhythm 8, longevity 100
- inputs: {"age_days": 2607, "days_push": 556, "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 1818, forks 389 (observed 2026-08-28T04:05:40.685353+00:00)

## What it is
A PyTorch library implementing multiple semantic segmentation models (DeepLab V3+, PSPNet, U-Net, SegNet, FCN, ENet, and others) with datasets, losses, LR schedulers, and data augmentation. It is configured via JSON config files and includes training and inference scripts plus a Colab tutorial.

## Use cases
- train a semantic segmentation model in pytorch
- implement deeplab v3+ for image segmentation
- segment images with u-net
- compare semantic segmentation architectures
- run inference with a pretrained segmentation model
- learn how semantic segmentation models are implemented

## When to choose
- you need a reference implementation of classic segmentation models in PyTorch
- you want a simple, config-driven training pipeline for segmentation research
- you are learning how models like DeepLab or PSPNet work internally

## When to avoid
- you need production-grade, actively maintained segmentation tooling like MMSegmentation or Detectron2
- you need the latest transformer-based segmentation models (SegFormer, Mask2Former)
- you need instance or panoptic segmentation rather than semantic segmentation

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, computer-vision
- domain: computer-vision, deep-learning, machine-learning, image-processing
- platform: python, cross-platform
- tags: semantic-segmentation, pytorch, deeplab, unet, pspnet, fcn, segnet, research-code, gpu

## Member repositories
- yassouali/pytorch-segmentation (main) score 26

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:40.685353+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 2026-08-30T03:20:09.505297+00:00, confidence not recorded.
  - readme: https://github.com/yassouali/pytorch-segmentation (fetched 2026-08-28T04:05:40.685353+00:00, sha 648aa5a6c171)
- Data as of 2026-08-30T08:39:29.467469+00:00.