jiaheliu
Results
2
issues of
jiaheliu
Thank you so much for this work. Currently, I'm trying to reproduce the results in the paper. I kept all the hyperparameters the same as the source code of this...
When sampling a segment of video to train the temporal module, the length of the dataset is divided by the `train_sample_interval = 4`. But in this line: https://github.com/Wangt-CN/DisCo/blob/8538889c9ee9edd8dd43ffee182d1a91ce7a9828/dataset/tiktok_video_dataset.py#L295 the code...