# megvii-research/ML-GCN

PyTorch implementation of Multi-Label Image Recognition with Graph Convolutional Networks, CVPR 2019.

Repository: https://github.com/megvii-research/ML-GCN
Canonical: https://ross.abutalabs.com/products/ml-gcn
Language: Python
License Family: other
Last push: 2021-12-31T11:51:53+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2737, "days_push": 1706, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1446, forks 346 (observed 2026-08-28T04:04:45.315547+00:00)

## What it is
A PyTorch implementation of ML-GCN, the CVPR 2019 paper 'Multi-Label Image Recognition with Graph Convolutional Networks'. It provides training and evaluation scripts with pretrained checkpoints for COCO, NUS-WIDE, and VOC 2007 multi-label image classification.

## Use cases
- classify images with multiple labels
- reproduce ML-GCN results from CVPR 2019
- train a multi-label image recognition model on COCO or VOC
- use graph convolutional networks for label dependency modeling
- evaluate pretrained multi-label classification checkpoints
- build a research baseline for multi-label recognition

## When to choose
- you need a reference implementation of ML-GCN for research or paper reproduction
- you want pretrained multi-label classification models for COCO or VOC 2007
- you are studying how GCNs can model label correlations in image recognition

## When to avoid
- you need a production-ready, actively maintained image classification system
- you require a permissive license - the repository has no license
- you need modern PyTorch versions - the code targets torch 0.3.1
- you need general single-label image classification

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, computer-vision
- domain: computer-vision, image-processing, machine-learning, deep-learning
- platform: python
- tags: pytorch, multi-label-classification, graph-convolutional-networks, gcn, cvpr-2019, research-code, image-recognition

## Member repositories
- megvii-research/ML-GCN (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:45.315547+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 2026-08-30T04:36:10.846433+00:00, confidence not recorded.
  - readme: https://github.com/megvii-research/ML-GCN (fetched 2026-08-28T04:04:45.315547+00:00, sha d56f4ef480a8)
- Data as of 2026-08-30T08:39:29.467469+00:00.