bbox_iou __call__() takes 2 positional arguments but 3 were given

[turboxin]

when training with custom data, got this error during evaluate:

Traceback (most recent call last): File "train.py", line 169, in batch_size=8, File "/data/PyTorch-YOLOv3/test.py", line 48, in evaluate outputs = non_max_suppression(outputs, conf_thres=conf_thres, nms_thres=nms_thres) File "/data/PyTorch-YOLOv3/utils/utils.py", line 286, in non_max_suppression large_overlap = bbox_iou(detections[0, :4].unsqueeze(0), detections[:, :4]) > nms_thres TypeError: call() takes 2 positional arguments but 3 were given

however only 2 arguments are given to the bbox_iou function:

detections[0, :4].unsqueeze(0)= tensor([[791.5812, 624.4221, 895.8517, 695.7702]]) detections[:, :4]= tensor([[791.5812, 624.4221, 895.8517, 695.7702], [788.4619, 623.7973, 914.6245, 696.2021], [778.1987, 627.6257, 924.9369, 707.6843], [778.9758, 627.8815, 908.5028, 707.6105], [795.1968, 618.4579, 892.1761, 686.6302], [791.0020, 617.0222, 912.0226, 687.8959], [774.2858, 631.4473, 929.0527, 719.4227], [831.1809, 625.9000, 952.0428, 693.0446], [782.9749, 622.2780, 935.9423, 697.6901], [771.0724, 623.4414, 947.9097, 711.9248], [839.4044, 626.1903, 959.7700, 692.9878], [779.0910, 626.6616, 892.4577, 693.1209], [764.6205, 628.5501, 906.9672, 706.4101], [786.1196, 621.8602, 885.3606, 682.8246], [775.5361, 622.6221, 863.6934, 697.6937], [776.5481, 625.6379, 878.6750, 694.2166], [767.1235, 622.3908, 856.2692, 697.9639], [509.9140, 634.5790, 522.4455, 670.5878], [517.4629, 634.6817, 530.1768, 670.0563]])