# Lam1360/YOLOv3-model-pruning

在 oxford hand 数据集上对 YOLOv3 做模型剪枝（network slimming）

Repository: https://github.com/Lam1360/YOLOv3-model-pruning
Canonical: https://ross.abutalabs.com/products/yolov3-model-pruning
Language: Python
License: MIT
License Family: permissive
Topics: yolov3, model-pruning, hand-detection, channel-pruning, object-detection
Last push: 2022-09-26T19:23:44+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2623, "days_push": 1437, "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 1676, forks 429 (observed 2026-08-28T04:05:20.538885+00:00)

## What it is
A PyTorch implementation of YOLOv3 channel pruning (network slimming) applied to hand detection on the Oxford Hand dataset. It provides sparse training, pruning, and fine-tuning scripts that shrink the model to about one-sixth the parameters and one-third the FLOPs while keeping mAP essentially unchanged.

## Use cases
- prune yolov3 to reduce model size and inference latency
- apply network slimming channel pruning to an object detector
- train a hand detector on the oxford hand dataset
- speed up yolov3 inference on gpu
- compress a detection model without losing accuracy
- learn how bn gamma based pruning works

## When to choose
- you need a smaller, faster YOLOv3 detector and can retrain or fine-tune on your own dataset
- you want a working reference implementation of network slimming for detection models
- your target is hand detection specifically and you want pretrained pruned weights

## When to avoid
- you need a maintained production-ready detection framework
- you want pruning for classification models as in the original paper
- you cannot run PyTorch 1.x training pipelines or lack GPU resources
- you expect the same compression ratios on datasets other than Oxford Hand

## 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
- tags: yolov3, model-pruning, network-slimming, channel-pruning, object-detection, hand-detection, pytorch, model-compression, sparse-training, oxford-hand-dataset, gpu

## Member repositories
- Lam1360/YOLOv3-model-pruning (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:20.538885+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 2026-08-30T03:42:25.472985+00:00, confidence not recorded.
  - readme: https://github.com/Lam1360/YOLOv3-model-pruning (fetched 2026-08-28T04:05:20.538885+00:00, sha 27049ee6fbb2)
- Data as of 2026-08-30T08:39:29.467469+00:00.