PyTorch-YOLOv3 train on voc dataset

I want to train on VOC ,and I modified voc.data,voc.names and yolo.cfg.And I use voc_label.py to get 2007_train.text and 2007_test.text.

When I run "python3 train.py --data_config config/voc.data --pretrained_weights weights/darknet53.conv.74",there are some errors:

Traceback (most recent call last): File "train.py", line 99, in for batch_i, (_, imgs, targets) in enumerate(dataloader): File "/home/yyx/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in next return self._process_next_batch(batch) File "/home/yyx/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) TypeError: Traceback (most recent call last): File "/home/yyx/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/yyx/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/yyx/PyTorch-YOLOv3-master/utils/datasets.py", line 130, in getitem img, targets = horisontal_flip(img, targets) File "/home/yyx/PyTorch-YOLOv3-master/utils/augmentations.py", line 8, in horisontal_flip targets[:, 2] = 1 - targets[:, 2] TypeError: 'NoneType' object is not subscriptable

Is there anything else that needs to be changed?

Jun 11 '19 11:06 yyx1107

I also has this problem, had you solve it?

Jun 18 '19 13:06 YingYingQinn

@YingYingQinn I haven't solved it yet. There are some opinions that others told me.Can you tell me if you solve this problem? 1)Check your images, some of the images may be broken, you can print the image path when come with this problem. Use try..... except...... @yyx1107 2)@yyx1107 It's my file path problem, I modified the txt file location.

Jun 19 '19 07:06 yyx1107