New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Which dataset is used for training? #7

Open

117kyjin opened this issue Sep 29, 2018 · 1 comment

117kyjin commented Sep 29, 2018 •

edited

Loading

Thank you for sharing caffe2pytorch code and trained models, but I have two questions.

Which dataset is used in training each model you provide? I think each model is trained by different dataset as following. Is it correct?

hmdb51_flow.pth: hmdb51 split 1 training set (optical flow only)
hmdb51_rgb.pth: hmdb51 split 1 training set (RGB only)
kinetics_flow.pth: kinetics split 1 training set (optical flow only)
kinetics_rgb.pth: kinetics split 1 training set (RGB only)
ucf101_flow.pth: ucf101 split 1 training set (optical flow only)
ucf101_rgb.pth: ucf101 split 1 training set (RGB only)

All models are pre-trained by ImageNet?

Owner

shuangshuangguo commented Oct 5, 2018

Yes, you are right both.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment