Hello~
When I was running "Python train.py -- gpu_idx 0-- batch_size 4-- num_workers 8", the speed was particularly slow, and I had to wait for about five minutes before the first epoch appeared. Later, when the first epoch training reached the 16th round, an error message appeared and the operation stopped. The error message is as follows:
Traceback (most recent call last):
File "train.py", line 250, in <module>
main()
File "train.py", line 52, in main
main_worker(configs.gpu_idx, configs)
File "train.py", line 149, in main_worker
train_one_epoch(train_dataloader, model, optimizer, lr_scheduler, epoch, configs, logger, tb_writer)
File "train.py", line 206, in train_one_epoch
total_loss, outputs = model(imgs, targets)
File "D:\Learning_Tool\AnacondaEP\envs\comyolo4\lib\site-packages\torch\nn\modules\module.py", line 550, in __call__
result = self.forward(*input, **kwargs)
File "D:\Learning_Tool\Pycharm\Project\Complex-YOLOv4-Pytorch-master\src\models\darknet2pytorch.py", line 221, in forward
x, layer_loss = self.models[ind](x, targets, img_size, self.use_giou_loss)
File "D:\Learning_Tool\AnacondaEP\envs\comyolo4\lib\site-packages\torch\nn\modules\module.py", line 550, in __call__
result = self.forward(*input, **kwargs)
File "D:\Learning_Tool\Pycharm\Project\Complex-YOLOv4-Pytorch-master\src\models\yolo_layer.py", line 197, in forward
pred_boxes=pred_boxes, pred_cls=pred_cls, target=targets, anchors=self.scaled_anchors)
File "D:\Learning_Tool\Pycharm\Project\Complex-YOLOv4-Pytorch-master\src\models\yolo_layer.py", line 135, in build_targets
GIoU=self.use_giou_loss)
File "D:\Learning_Tool\Pycharm\Project\Complex-YOLOv4-Pytorch-master\src\utils\iou_rotated_boxes_utils.py", line 119, in iou_pred_vs_target_boxes
p_poly, t_poly = cvt_box_2_polygon(p_cons), cvt_box_2_polygon(t_cons)
File "D:\Learning_Tool\Pycharm\Project\Complex-YOLOv4-Pytorch-master\src\utils\iou_rotated_boxes_utils.py", line 31, in cvt_box_2_polygon
return Polygon([(box[i, 0], box[i, 1]) for i in range(len(box))]).buffer(0)
File "D:\Learning_Tool\AnacondaEP\envs\comyolo4\lib\site-packages\shapely\geometry\polygon.py", line 243, in __init__
ret = geos_polygon_from_py(shell, holes)
File "D:\Learning_Tool\AnacondaEP\envs\comyolo4\lib\site-packages\shapely\geometry\polygon.py", line 509, in geos_polygon_from_py
ret = geos_linearring_from_py(shell)
File "shapely\speedups\_speedups.pyx", line 408, in shapely.speedups._speedups.geos_linearring_from_py
ValueError: GEOSGeom_createLinearRing_r returned a NULL pointer
I don't know the reason for this, could you please provide suggestions? Thank you!
Hello~ When I was running "Python train.py -- gpu_idx 0-- batch_size 4-- num_workers 8", the speed was particularly slow, and I had to wait for about five minutes before the first epoch appeared. Later, when the first epoch training reached the 16th round, an error message appeared and the operation stopped. The error message is as follows:
I don't know the reason for this, could you please provide suggestions? Thank you!