-
Notifications
You must be signed in to change notification settings - Fork 0
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
RASR FSA Builder #59
RASR FSA Builder #59
Conversation
I'm not sure about a test for this. It does require the librasr package, which I'm not sure if we should put it into the requirements and to initialize the |
You "could" add extra dependencies just for a test, including a small dummy lexicon and corpus for testing. While this would be preferable to have, I am not sure if this is within your time-scope. |
I would like to do it, actually. Having a small dummy corpus and lexicon is a good idea. However, the installation of |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Excellent documentation, thank you!
i6_models/parts/fsa.py
Outdated
fsa by simple left-multiplication and moving the tensors to a different device. | ||
It can simply be passed to `i6_native_ops.fbw.fbw_loss` and `i6_native_ops.fast_viterbi.align_viterbi`. | ||
:param num_states: the total number of all states S | ||
:param edges: a [4, E] tensor of edges where each column is an edge consisting |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nitpick: E
(I guess number of edges) is undefined here.
Co-authored-by: michelwi <[email protected]>
Co-authored-by: Albert Zeyer <[email protected]>
Co-authored-by: Albert Zeyer <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
lgtm now. Would you give it another quick test before merging?
Components for building FSAs with the
librasr
python package and using them in a PyTorch training