jwyang/fpn.pytorch
jwyang/fpn.pytorch has 969 GitHub stars and 220 forks. GitHot measured 0 stars in 24 hours. Last checked 2026-11-19 13:41 UTC. Last push 2019-01-11.
Gemessener Sternzuwachs
- in 24h
- 0
- in 7 days
- Verlauf wird gesammelt
- in 15 days
- Verlauf wird gesammelt
- in 1 month
- Verlauf wird gesammelt
Nettoänderungen in überlappenden Zeiträumen, keine Tagessummen. Fehlende Messwerte bleiben leer. Sterne zeigen Interesse, nicht Qualität.
Über uns
Pytorch implementation of Feature Pyramid Network (FPN) for Object Detection
Maschinelles LernenREADME.md
Originalfpn.pytorch Pytorch implementation of Feature Pyramid Network (FPN) for Object Detection
Introduction
This project inherits the property of our pytorch implementation of faster r-cnn. Hence, it also has the following unique features:
It is pure Pytorch code. We convert all the numpy implementations to pytorch.
It supports trainig batchsize > 1. We revise all the layers, including dataloader, rpn, roi-pooling, etc., to train with multiple images at each iteration.
It supports multiple GPUs. We use a multiple GPU wrapper (nn.DataParallel here) to make it flexible to use one or more GPUs, as a merit of the above two features.
It supports three pooling methods. We integrate three pooling methods: roi pooing, roi align and roi crop. Besides, we convert them to support multi-image batch training.
Benchmarking
We benchmark our code thoroughly on three datasets: pascal voc, coco. Below are the results:
1). PASCAL VOC 2007 (Train/Test: 07trainval/07test, scale=600, ROI Align)
| model | GPUs | Batch Size | lr | lr_decay | max_epoch | Speed/epoch | Memory/GPU | mAP |
|---|---|---|---|---|---|---|---|---|
| Res-101 | 8 TitanX | 24 | 1e-2 | 10 | 12 | 0.22 hr | 9688MB | 74.2 |
Results on coco are on the way.