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

About training. #15

Open

hirusnov opened this issue Nov 1, 2020 · 4 comments

hirusnov commented Nov 1, 2020

How many times do you train to get the best results?
Can I use the old pre-trained model to continue train new models?

Owner

HongguLiu commented Nov 2, 2020

We usually train our model in 50-100 epochs.
Of course, you can use the pretraiend model and finetune for your task.

Author

hirusnov commented Nov 2, 2020

Ohhhh, I saw you use 40 videos original and 40 videos deepfake with 50 100 epochs, is it right?

Owner

HongguLiu commented Nov 2, 2020

We use 720 videos and 140 videos to train and validate, and we randomly extract 50 frames for each video.

Author

hirusnov commented Nov 2, 2020

Okayyy, thank you so much.
I understand

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