# tusen-ai/simpledet

A Simple and Versatile Framework for Object Detection and Instance Recognition

Repository: https://github.com/tusen-ai/simpledet
Canonical: https://ross.abutalabs.com/products/simpledet
Language: Python
License: Apache-2.0
License Family: permissive
Topics: mxnet, object-detection, instance-segmentation
Last push: 2021-09-23T02:57:09+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": 2773, "days_push": 1805, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3085, forks 481 (observed 2026-08-28T04:07:42.069299+00:00)

## What it is
SimpleDet is a Python framework built on MXNet for object detection and instance recognition. It provides state-of-the-art detection models (FasterRCNN, MaskRCNN, RetinaNet, TridentNet, NASFPN, EfficientNet, etc.) with FP16 training and scalable distributed training out of the box.

## Use cases
- train object detection models on COCO
- run instance segmentation with Mask R-CNN
- train detection models with FP16 mixed precision on GPU
- distributed multi-GPU object detection training
- benchmark state-of-the-art detection architectures like TridentNet and NASFPN
- fine-tune pretrained detection models on custom datasets
- knowledge distillation for detection models

## When to choose
- you need a modular, config-driven detection framework on MXNet
- you want memory-efficient FP16 or INT8 training for detection
- you need out-of-the-box scalable distributed training
- you want broad coverage of detection and instance segmentation models in one codebase

## When to avoid
- your stack is PyTorch or TensorFlow based
- you need actively maintained software - the last release was 2021
- you only need inference without training
- you work outside Linux/Ubuntu environments without Docker

## Facets
- artifact type: framework
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing, gpu-computing
- domain: computer-vision, deep-learning, machine-learning, image-processing
- platform: python
- tags: object-detection, instance-segmentation, mxnet, fp16-training, distributed-training, mask-rcnn, faster-rcnn, retinanet, knowledge-distillation, linux, gpu, docker

## Member repositories
- tusen-ai/simpledet (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:42.069299+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 2026-08-30T07:27:40.180751+00:00, confidence not recorded.
  - readme: https://github.com/tusen-ai/simpledet (fetched 2026-08-28T04:07:42.069299+00:00, sha 975e4ba38e99)
- Data as of 2026-08-30T08:39:29.467469+00:00.