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

read me for this project #1

Open
Damanotra opened this issue Jan 20, 2021 · 1 comment
Open

read me for this project #1

Damanotra opened this issue Jan 20, 2021 · 1 comment

Comments

@Damanotra
Copy link

Hi, thank you for building this project

It seems the readme is for JoeyNMT instead of your actual sigmorphon project. May I know how to get hands on your model (inference, training. etc)? Thank you

@bpopeters
Copy link
Collaborator

Hello, thank you for your interest and sorry for the incredibly long time in replying. I'll update the readme tomorrow, after running a few tests to make sure things really work the way I think they do. The gist of using it is that everything is specified in config files, of which I've included several (I hope still) working examples in configs/. Training a multilingual g2p model is accomplished by running

python -m joeynmt train configs/task1/multi/rnn/multi-128-512-2-2-langall.yaml

The only thing you should need to change about this config is the path to the data -- you can replace those lines with your own. If memory serves, you can manually specify all of the paths of the files, or use glob syntax (as in the provided example). I think the code will parse the language code from the name of the file, so it's important not to rename any of the files.

At test time, you can run

python -m joeynmt test configs/task1/multi/rnn/multi-128-512-2-2-langall.yaml

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