v0.1.0
Pre-release
Pre-release
- Realized the YOLOv2 network and YOLOv2 loss function definition;
- Implemented complete training/evaluation and prediction code;
- Darknet19 and FastDarknet19 use ImageNet for pretraining, with an input size of 224x224;
- Train using voc07+12 training data, evaluate based on voc2007 test.
| Original (darknet) | tztztztztz/yolov2.pytorch | zjykzj/YOLOv2(This) | |
|---|---|---|---|
| train | VOC2007+2012 trainval | VOC2007+2012 trainval | VOC2007+2012 trainval |
| val | VOC2007 test | VOC2007 test | VOC2007 test |
| VOC AP[IoU=0.50], inference | 76.8 | 72.7 | 69.69 |
| conf_thre | / | 0.005 | 0.005 |
| nms_thre | / | 0.45 | 0.45 |
| input_size | 416 | 416 | 416 |