PyTorchConv3D icon indicating copy to clipboard operation
PyTorchConv3D copied to clipboard

FileNotFoundError: n_frames_file_path does not exist

Open lhyfst opened this issue 4 years ago • 1 comments

When I tried to run the example

python train.py --dataset=ucf101 --model=resnet --video_path=/home/tomrunia/data/UCF-101/jpg --annotation_path=/home/tomrunia/data/UCF-101/ucfTrainTestlist/ucf101_01.json --batch_size=64 --num_classes=101 --momentum=0.9 --weight_decay=1e-3 --model_depth=34 --resnet_shortcut=A --spatial_size=112 --sample_duration=16 --optimizer=SGD --learning_rate=0.01

there is an error:

...
############################################################
Data normalization: use dataset mean, no dataset std
[Thursday 10:42] Preparing datasets...
Loading UCF-101 videos [0/9537]
Traceback (most recent call last):
  File "train.py", line 121, in <module>
    data_loaders = data_factory.get_data_loaders(config, train_transforms, validation_transforms)
  File "/home/heyuan/Research/PyTorchConv3D/factory/data_factory.py", line 201, in get_data_loaders
    train_transforms['temporal'], train_transforms['target'])
  File "/home/heyuan/Research/PyTorchConv3D/factory/data_factory.py", line 52, in get_training_set
    target_transform=target_transform)
  File "/home/heyuan/Research/PyTorchConv3D/datasets/ucf101.py", line 176, in __init__
    sample_duration)
  File "/home/heyuan/Research/PyTorchConv3D/datasets/ucf101.py", line 106, in make_dataset
    raise FileNotFoundError('n_frames_file_path does not exist: {}'.format(n_frames_file_path))
FileNotFoundError: n_frames_file_path does not exist: /home/heyuan/Datasets/ucf101/jpg/ApplyEyeMakeup/v_ApplyEyeMakeup_g08_c01/n_frames

Does anybody know how to solve this problem? Thanks in advance!

lhyfst avatar May 07 '20 17:05 lhyfst

I get the same problem,can you tell me how to solve it

zhaoleo1111 avatar Nov 19 '20 06:11 zhaoleo1111