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

Training on custom MIDI corpus? #2

Closed
robinsloan opened this issue Feb 4, 2020 · 5 comments
Closed

Training on custom MIDI corpus? #2

robinsloan opened this issue Feb 4, 2020 · 5 comments
Labels
enhancement New feature or request

Comments

@robinsloan
Copy link

Hello! This is a great project, and I was able to get it running immediately. Thank you for sharing it!

Is there a way to train the model from scratch -- and/or fine-tune one of your published models -- on my own corpus of MIDI files?

@remyhuang
Copy link
Contributor

Hi! I have uploaded the finetune.py.
Please help me try it and tell me if there is anything wrong!

Thanks,
Remy

@remyhuang
Copy link
Contributor

remyhuang commented Feb 5, 2020

I would like to save every checkpoint per epoch for now. So, you have to choose which checkpoint you like the most to further generate samples. Basically, you can try every checkpoint and listen to the generated output.

In my personal experiments, by using ~30 minutes midi to finetune on one V100, it only took 5-15 minutes to obtain good generated examples.

However, it is still based on your personal data.

Thanks,
Remy

@robinsloan
Copy link
Author

It works perfectly! Incredible! Thank you so much for the quick reply & the helpful addition. This is great!

@remyhuang remyhuang added the enhancement New feature or request label Feb 13, 2020
@huangqing6
Copy link

Hello, I was wondering how I can turn my midi file into a file inside REMI-tempo-checkpoint?

@dedededefo
Copy link

它完美地工作!极好的!非常感谢您的快速回复和有用的补充。这很棒!

Hello, can you tell me how to use your MIDI fine-tuning

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

No branches or pull requests

4 participants