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

preprocess data for "Dual2seq" model #3

Open
Lavine24 opened this issue Jun 28, 2019 · 3 comments
Open

preprocess data for "Dual2seq" model #3

Lavine24 opened this issue Jun 28, 2019 · 3 comments

Comments

@Lavine24
Copy link

Hi,
Thanks for the awesome work, I meet some questions on preprocess the data.
1. I use your release data in nc-v11, but find no "nc-v11.tok_le50.json" and "vectors.en.st" file on the release data.
2. I try to use the dual_to_seq/data scrips to generate the json file and use the pre-trained embedding file. are this the true preprocess methods on the "dual2seq" model?
Thank you very much!

@freesunshine0316
Copy link
Owner

  1. Yes they are not included and you have to generate them by running the provided scripts.
  2. Yes.

@lidongxing
Copy link

@freesunshine0316
Where is the pretrained embeddings of 'de' language from or trained from scratch with some corpus? Tks.

@freesunshine0316
Copy link
Owner

@lidongxing The embeddings of 'de' are jointly trained from scratch during NMT training.

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

3 participants