# experiencor/keras-yolo3

Training and Detecting Objects with YOLO3

Repository: https://github.com/experiencor/keras-yolo3
Canonical: https://ross.abutalabs.com/products/experiencor-keras-yolo3
Language: Python
License: MIT
License Family: permissive
Topics: yolo, deep-learning, object-detection
Last push: 2023-09-05T07:29:29+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": 3069, "days_push": 1093, "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 1608, forks 838 (observed 2026-08-28T04:05:10.768108+00:00)

## What it is
A Keras/TensorFlow implementation of YOLOv3 for object detection, supporting detection with pretrained weights, custom model training with warmup and multi-scale techniques, multi-GPU training, and mAP evaluation. It includes a model zoo with pretrained configurations for datasets like kangaroo, raccoon, license plate, blood cell, and VOC detection.

## Use cases
- train a custom YOLOv3 object detector on my own dataset
- run object detection on images with pretrained YOLOv3 weights
- evaluate object detection model mAP on VOC-style datasets
- train YOLOv3 on multiple GPUs
- detect license plates or animals in images
- convert VOC-format annotations for YOLO training

## When to choose
- you need a pure Keras/TensorFlow YOLOv3 implementation you can read and modify
- you want to train a custom single-class or few-class detector on a small dataset
- you need mAP evaluation alongside training in one codebase
- you want multi-GPU YOLOv3 training in Keras

## When to avoid
- you need the latest YOLO versions (YOLOv5/v8) or state-of-the-art accuracy
- you need COCO evaluation, which is not yet implemented
- you want production-grade maintenance and recent TensorFlow compatibility
- you need alternative backbones like MobileNet, ResNet, or DenseNet, which are listed as unfinished

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, machine-learning, image-processing
- platform: python, cross-platform
- tags: yolo, object-detection, keras, tensorflow, training, inference, map-evaluation, voc-dataset, gpu

## Member repositories
- experiencor/keras-yolo3 (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:10.768108+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 2026-08-30T03:51:25.448261+00:00, confidence not recorded.
  - readme: https://github.com/experiencor/keras-yolo3 (fetched 2026-08-28T04:05:10.768108+00:00, sha d0308d0d7d53)
- Data as of 2026-08-30T08:39:29.467469+00:00.