yolov5-face icon indicating copy to clipboard operation
yolov5-face copied to clipboard

YOLO5Face: Why Reinventing a Face Detector (https://arxiv.org/abs/2105.12931) ECCV Workshops 2022)

Results 155 yolov5-face issues
Sort by recently updated
recently updated
newest added

Is there any way I can train a pretrained model like 'yolov5s-face.pt' using the train.py script from this repo on a custom dataset? I would like to conduct an experiment...

When I run the 'test_widerface.py' file, the following error appears. Namespace(agnostic_nms=False, augment=False, classes=None, conf_thres=0.02, dataset_folder='../WiderFace/val/images/', device='0', exist_ok=False, folder_pict='/yolov5-face/data/widerface/val/wider_val.txt', img_size=640, iou_thres=0.5, name='exp', project='runs/detect', save_folder='./widerface_evaluate/widerface_txt/', update=False, weights=['./weights/download_weights.sh']) Traceback (most recent call last):...

Scanning '/home1/data/face_data/train' for images and labels... 98626 found, 8694 missing, 0 empty, 79 corrupted: 这个是啥意思

楼主,你好,非常您开源这个前沿的工作。我在使用您提供的yolov5l-face,pt进行转trt的验证时发现耗时对不上。(CDOD) root@workspace:/dfs/data/pCode/yolov5_face# python torch2trt/speed.py --torch_path ./yolov5l-face.pt --trt_path ./yolov5l-face.trt Fusing layers... /dfs/data/anaconda/envs/CDOD/lib/python3.8/site-packages/torch/functional.py:445: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:2157.)...

训练完成的yolov5s的模型的map只有0.324(@0.5:0.95)精度正常吗

您好,这个的YOLOv5n-0.5 | ShuffleNetv2 | 90.76 | 88.12 | 73.82 | 0.447 | 0.571配置文件可以分享一下吗,我按照YOLOv5n-0.5.yaml进行训练,模型和这个不太一样,

大佬您好,我再训练数据的时候,Class Images Targets P R [email protected] [email protected]:.95: 100%|,这些值都是0,这个该怎么解决啊 @djmmoss @muhammad-tayyab @hylrh2008 @bobo0810 @wjtan99

yolov5-face-master/models/yolo.py", line 67, in forward y[..., class_range] = x[i][..., class_range].sigmoid() IndexError: index 85 is out of bounds for dimension 0 with size 85 Whats the problem? how to solve it?