Skip to content

Conversation

lpizzinidev
Copy link
Contributor

No description provided.

Copy link
Contributor

@fchollet fchollet left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the PR!

This example will need more changes, for instance there are tf.keras references remaining. I think the whole example can be made backend-agnostic (with preprocessing in tf.data).

examples/nlp/.DS_Store

Please revert this file.

@lpizzinidev lpizzinidev force-pushed the multi-label-classification-v3 branch from 9659c55 to 2d76b3c Compare March 5, 2024 18:37
@lpizzinidev
Copy link
Contributor Author

@fchollet
Thanks for the feedback! (and sorry for messing up the files)
Maybe I'm missing something, but I don't think that ragged tensors are currently supported by Keras v3 so I'm not sure we can make this backend-agnostic.

"""

"""
## Imports
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you update Last modified above and add yourself to the authors list?

layer.
"""

terms = tf.ragged.constant(train_df["terms"].values)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Would it work to use padding here instead of ragged tensors?
https://keras.io/api/ops/numpy/#pad-function

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

Successfully merging this pull request may close these issues.

4 participants