123liluky

print redundant lines while training

4

Whiel I was training, lines besides epoch and loss were printed redundantly like below.  Do you know how to avoid this?

images used in train

2

In UCF101_ResNetCRNN.py: begin_frame, end_frame, skip_frame = 1, 29, 1 selected_frames = np.arange(begin_frame, end_frame, skip_frame).tolist() train_set, valid_set = Dataset_CRNN(data_path, train_list, train_label, selected_frames, transform=transform), Dataset_CRNN(data_path, test_list, test_label, selected_frames, transform=transform) So, you just...

setup problem

2

Thanks for your job. Must we install tf1.6 instead of any other versions? And what version of cuda and cudnn should I install? I have installed tf1.4, cuda8.0 and cudnn6.

Do you know any convenient temporal annotation tool?Thank you.

overfitting

1

I used my own small dataset to extract I3D features. Then use main.py to train. Overfitting happened in training set. Do you have any suggestions? 

I can't clone the repository successfully. How can I do? 

Do you know any convenient temporal annotation tool?Thank you.

Which temporal annotation tool did you use?Thank you.

Do you know any convenient temporal annotation tool?Thank you.

train BrnoCompSpeed

8

I commented lines 335-366 and uncommented lines 315-333, then ran dataset_utils/fromBrnoCompSpeed.py to produce the training data. NameError happens : name 'pkl_paths' is not defined.