data-path and code training error

[obinnadyke]

Hi!
I tried to train the model with UCF101 dataset and test on HMDB51 but i got the errors. After debugging the below error indicates model cannot read the data. Please confirm that the model can train video data because I noticed that your 'cropped nec-drone' data is images only. Also pls describe how to structure the data-path/directory.
I used the following:

python3 main.py \
    --train_source_dataset ./data/train/source \
    --val_dataset ./data/val/target \
    --epochs 60 \
    --optimizer sgd \
    --lr 0.001 \
    --weight_decay 1e-9 \
    --scheduler cosine \
    --batch_size 32 \
    --n_clips 1 \
    --n_frames 16 \
    --frame_size 224 \
    --num_workers 4 \
    --gpus 0 \
    --train head+partial \
    --mlp_hidden_dim 2048 \
    --mlp_n_layers 0 \
    --replace_with_mlp \
    --name hmdb-ucf-source-head+partial \
    --project transformer_da \
    --save_model \
    --wandb

2. I got the following error:

Global seed set to 5
Traceback (most recent call last):
File "main.py", line 183, in
main()
File "main.py", line 141, in main
num_workers=args.num_workers,
File "/home/User/Desktop/UDAVT/src/dataloader.py", line 275, in prepare_data
num_workers=num_workers,
File "/home/user/Desktop/UDAVT/src/dataloader.py", line 237, in prepare_dataloaders
drop_last=True,
File "/home/User/anaconda3/envs/venv/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 344, in init
sampler = RandomSampler(dataset, generator=generator) # type: ignore[arg-type]
File "/home/user/anaconda3/envs/venv/lib/python3.7/site-packages/torch/utils/data/sampler.py", line 108, in init
"value, but got num_samples={}".format(self.num_samples))
ValueError: num_samples should be a positive integer value, but got num_samples=0

Thanks as i wait your reply. I look forward to train and test your code on UCF101 and HMDB51. Give us link where to download hmdb<-->ucf video dataset too!