Tutorial-about-3D-convolutional-network icon indicating copy to clipboard operation
Tutorial-about-3D-convolutional-network copied to clipboard
verified publisher icon OValery16

What should I do?

Open Lihongjian520 opened this issue 6 years ago • 3 comments

usage: train.py [-h] [--config CONFIG] [--eval_only EVAL_ONLY] [--resume RESUME] [--gpus GPUS] [--optimizer OPTIMIZER]

PyTorch Jester Training using JPEG

optional arguments: -h, --help show this help message and exit --config CONFIG, -c CONFIG json config file path --eval_only EVAL_ONLY, -e EVAL_ONLY evaluate trained model on validation data. --resume RESUME, -r RESUME resume training from given checkpoint. --gpus GPUS, -g GPUS gpu ids for use. --optimizer OPTIMIZER, -o OPTIMIZER optimizer of choice

Can you give me some?

Lihongjian520 avatar Apr 18 '19 08:04 Lihongjian520

Traceback (most recent call last): File "E:/3dcnn/hdd/train.py", line 345, in main() File "E:/3dcnn/hdd/train.py", line 198, in main train_loss, train_top1, train_top5 = train( train_loader, model, criterion, optimizer, epoch) File "E:/3dcnn/hdd/train.py", line 230, in train for i, (input, target) in enumerate(train_loader): File "D:\python3\lib\site-packages\torch\utils\data\dataloader.py", line 336, in next return self._process_next_batch(batch) File "D:\python3\lib\site-packages\torch\utils\data\dataloader.py", line 357, in _process_next_batch raise batch.exc_type(batch.exc_msg) IndexError: Traceback (most recent call last): File "D:\python3\lib\site-packages\torch\utils\data\dataloader.py", line 106, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "D:\python3\lib\site-packages\torch\utils\data\dataloader.py", line 106, in samples = collate_fn([dataset[i] for i in batch_indices]) File "E:\3dcnn\hdd\data_loader.py", line 38, in getitem img_paths = self.get_frame_names(item.path) File "E:\3dcnn\hdd\data_loader.py", line 73, in get_frame_names frame_names += [frame_names[-1]] * (num_frames_necessary - num_frames) IndexError: list index out of range

Lihongjian520 avatar Apr 18 '19 14:04 Lihongjian520

For those whose os is windows, replace frame_names.extend(glob.glob(os.path.join(path, "*" + ext))) to frame_names.extend(glob.glob(path + "/*" + ext)) in data_loader.py's line 59 and line 123.

Also, replace

"train_data_folder": "/20bn-jester-v1/",
"val_data_folder": "/20bn-jester-v1/",

to

"train_data_folder": "20bn-jester-v1",
"val_data_folder": "20bn-jester-v1",

in config.json's line 5 and line 6.

moon5756 avatar Mar 05 '20 03:03 moon5756

Thanks for your help.

OValery16 avatar Mar 05 '20 16:03 OValery16