# osmr/imgclsmob

Sandbox for training deep learning networks

Repository: https://github.com/osmr/imgclsmob
Canonical: https://ross.abutalabs.com/products/imgclsmob
Language: Python
License: MIT
License Family: permissive
Topics: machine-learning, deep-learning, mxnet, gluon, pytorch, classification, imagenet, neural-network, image-classification, chainer, keras, tensorflow, pretrained-models, cifar, segmentation, tensorflow2, semantic-segmentation, human-pose-estimation, 3d-face-reconstruction
Last push: 2024-09-06T09:38:46+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2977, "days_push": 726, "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 3016, forks 551 (observed 2026-08-28T04:07:37.978916+00:00)

## What it is
A research sandbox providing (re)implementations of numerous deep learning computer vision models for classification, segmentation, detection, and pose estimation across six frameworks (MXNet/Gluon, PyTorch, Chainer, Keras, TensorFlow 1.x/2.x). It includes training/evaluation/conversion scripts and automatically loadable pretrained weights distributed via per-framework PyPI packages like pytorchcv.

## Use cases
- find pretrained image classification models for pytorch
- train a model on imagenet or cifar
- get the same model implemented in multiple frameworks
- semantic segmentation models with pretrained weights
- human pose estimation model implementations
- convert models between deep learning frameworks
- benchmark cnn architectures on imagenet

## When to choose
- you need a wide zoo of CNN architectures with pretrained weights in PyTorch, TensorFlow, Keras, Gluon, Chainer, or TF2
- you want consistent cross-framework implementations of the same model
- you need training and evaluation scripts for ImageNet, CIFAR, Pascal VOC, ADE20K, Cityscapes, or COCO

## When to avoid
- you need transformer-based or latest SOTA vision models only
- you want a high-level production inference API rather than model definitions
- you work outside the six supported frameworks

## Facets
- artifact type: library
- maturity: active
- function: deep-learning, machine-learning, image-processing, computer-vision
- domain: deep-learning, computer-vision, machine-learning, image-processing
- platform: python, cross-platform
- tags: pretrained-models, image-classification, semantic-segmentation, pose-estimation, pytorch, tensorflow, mxnet, keras, chainer, model-zoo, imagenet

## Member repositories
- osmr/imgclsmob (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:37.978916+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 2026-08-30T07:30:04.096139+00:00, confidence not recorded.
  - readme: https://github.com/osmr/imgclsmob (fetched 2026-08-28T04:07:37.978916+00:00, sha ae3a13388b87)
- Data as of 2026-08-30T08:39:29.467469+00:00.