Skip to content
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.

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

How to train the model? #4

Open
framsc opened this issue Jul 15, 2020 · 3 comments
Open

How to train the model? #4

framsc opened this issue Jul 15, 2020 · 3 comments

Comments

@framsc
Copy link

framsc commented Jul 15, 2020

Could you please share the training script? I can not reproduce the results after reading
FaSNet, https://doi.org/10.1109/ASRU46091.2019.9003849
TAC, https://doi.org/10.1109/ICASSP40776.2020.9054177

Thanks in advance.

@yluo42
Copy link
Owner

yluo42 commented Oct 15, 2020

Sorry for my late reply as I missed this issue. I will probably prepare one script for training by early Nov.

On the other hand, some recent literatures have used the FaSNet+TAC model as baselines (e.g. https://arxiv.org/abs/2010.01703 and https://arxiv.org/abs/2010.06007), so the models should be able to be reproduced.

@framsc
Copy link
Author

framsc commented Oct 15, 2020

Thanks for the reply. I have read these papers too. It seems that the researchers are quite interested in FaSNet and TAC. Maybe you can provide a pre-trained model thus we have the same model to talk about. So are the best hyper-parameters, such as batch size, learning rate.

Nice job. Hope to hear from you soon.

@ghost
Copy link

ghost commented Aug 11, 2024

How to train the fasnet model after getting the dataset generated.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants