# WongKinYiu/ScaledYOLOv4

Scaled-YOLOv4: Scaling Cross Stage Partial Network

Repository: https://github.com/WongKinYiu/ScaledYOLOv4
Canonical: https://ross.abutalabs.com/products/scaledyolov4
Language: Python
License: GPL-3.0
License Family: copyleft
Topics: machine-learning, deep-learning, ml, pytorch, yolo, object-detection, yolov3, yolov4, scaled-yolov4, yolov4-tiny, yolov4-csp, yolov4-large
Last push: 2024-11-03T00:36:25+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2118, "days_push": 669, "days_rel": 669, "gap_med": null, "n_releases_24m": 1}
- flags: prerelease_only
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2024, forks 563 (observed 2026-08-28T04:06:06.651868+00:00)

## What it is
PyTorch implementation of Scaled-YOLOv4, which scales the Cross Stage Partial Network (CSP) across model sizes from tiny to P7 for object detection. It provides pretrained weights, training, and evaluation code for YOLOv4-tiny, YOLOv4-CSP, and YOLOv4-large variants.

## Use cases
- train a custom YOLOv4 object detection model
- run real-time object detection on images and video
- detect small objects with high-resolution YOLOv4-P6/P7 models
- deploy a lightweight YOLOv4-tiny detector on edge devices
- benchmark YOLOv4 accuracy on COCO

## When to choose
- you need state-of-the-art YOLOv4-family detectors with pretrained COCO weights
- you want scalable detection models from tiny to very large (P5-P7)
- you are reproducing the Scaled-YOLOv4 paper results

## When to avoid
- you want the actively maintained YOLOv5/YOLOv7/YOLOv8 ecosystem instead
- you need non-GPL licensing (project is GPL-3.0)
- you need classification or segmentation rather than detection

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, machine-learning
- platform: python
- tags: object-detection, yolo, pytorch, scaled-yolov4, yolov4-csp, yolov4-tiny, real-time-detection, gpu, linux

## Member repositories
- WongKinYiu/ScaledYOLOv4 (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:06.651868+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 2026-08-30T02:59:48.904657+00:00, confidence not recorded.
  - readme: https://github.com/WongKinYiu/ScaledYOLOv4 (fetched 2026-08-28T04:06:06.651868+00:00, sha 30eb3382b30e)
- Data as of 2026-08-30T08:39:29.467469+00:00.