Add a param-free classification model #821
Conversation
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
tests/test_cls.py
Outdated
| ("นี่เป็นบทความหนึ่ง", "Neutral") | ||
| ] | ||
| model = GzipModel(training_data) | ||
| self.assertIsNotNone(model.predict("รู้สึกดีจัง", k=3)) |
There was a problem hiding this comment.
would it better to modify this assert to also verify the correctness of the prediction?
There was a problem hiding this comment.
also, I get the result from the test:
There was a problem hiding this comment.
I don't think we can expect much from this few-shot classifier. May be sentiment analysis is not the best application to test either.
There was a problem hiding this comment.
News classification seems to work well (at least on the paper)
There was a problem hiding this comment.
I don't think we can expect much from this few-shot classifier. May be sentiment analysis is not the best application to test either.
I generally agree, but I just wanted to make the test suite slightly less confusing.
|
Kudos, SonarCloud Quality Gate passed!
|
|
@wannaphong do we have to fix the MacOS error? |
I will fix the MacOS error after this pull request. The error is come from tensorflow (again). |
What does this changes
Add a param-free classification model from “Low-Resource” Text Classification: A Parameter-Free Classification Method with Compressors (Jiang et al., Findings 2023)
Your checklist for this pull request
🚨Please review the guidelines for contributing to this repository.