YOLOV3 使用 pytorch 实现，使用 pytorch 方法来定义网络 .zip

## YOLO v3 implementation With pytorch > this repository only contain the detection module and we don't need the cfg from original YOLOv3, we implement it with pytorch. This repository is based on the official code of [YOLOv3](https://github.com/pjreddie/darknet) and [pytorch-yolo-v3](https://github.com/ayooshkathuria/pytorch-yolo-v3). There's also an implementation for YOLOv3 already for pytorch, but it uses a config file rather than a normal pytorch approch to defining the network. One of the goals of this repository is to remove the cfg file. ## Requirements * Python 3.5 * OpenCV * PyTorch 0.4 ## Installation * Install PyTorch-0.4.0 by selecting your environment on the website and running the appropriate command. * Clone this repository * Compile the nms * convert yolov3.weights to pytorch ```shell cd YOLOv3_Pytorch ./make.sh mkdir weights cd weights wget https://pjreddie.com/media/files/yolov3.weights cd .. python convert_darknet.py --version coco --weights ./weights/yolov3.weights --save_name ./weights/convert_yolov3_coco.pth # we will get the convert_yolov3_coco.pth ``` ## Train > We only train voc dataset because we don't have enough gpus to train coco datatset. This is still an experimental repository, we don't reproduce the original results very well. ### dataset [merge VOC dataset](https://github.com/yqyao/DRFNet#voc-dataset) * structure ./data/datasets/VOCdevkit0712/VOC0712/Annotations ./data/datasets/VOCdevkit0712/VOC0712/ImageSets ./data/datasets/VOCdevkit0712/VOC0712/JPEGImages * COCO Same with [COCO](https://github.com/yqyao/DRFNet#coco-dataset) ### train > you can train multiscale by changing data/config voc_config multiscale * convert weights ```shell cd weights wget wget https://pjreddie.com/media/files/darknet53.conv.74 cd ../ python convert_darknet.py --version darknet53 --weights ./weights/darknet53.conv.74 --save_name ./weights/convert_darknet53.pth ``` * train yolov3 ```python python train.py --input_wh 416 416 -b 64 --subdivisions 4 -d VOC --basenet ./weights/convert_darknet53.pth ``` ### eval ```python python eval.py --weights ./weights/convert_yolov3_voc.pth --dataset VOC --input_wh 416 416 ``` > darknet voc is trained by darknet, pytorch voc is trained by this repository **results** | darknet voc 608 | darknet voc 416 | pytorch voc 608| pytorch voc 416| |:-: |:-: | :-: |:-: | | 77.2 % | 76.2% | 74.7% | 74.9% | | 27ms | 18ms | 27ms | 18ms | ## Demo ```python python demo.py --images images --save_path ./output --weights ./weights/convert_yolov3_coco.pth -d COCO ``` ## Example <img align="center" src= "https://github.com/yqyao/YOLOv3_Pytorch/blob/master/output/output_person.jpg"> <!--  --> ## References - [YOLOv3: An Incremental Improvemet](https://pjreddie.com/media/files/papers/YOLOv3.pdf) - [Original Implementation (Darknet)](https://github.com/pjreddie/darknet) - [pytorch-yolo-v3](https://github.com/ayooshkathuria/pytorch-yolo-v3) - [pytorch-yolo2](https://github.com/marvis/pytorch-yolo2)