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

[MRG] Fix FastText word-vectors w/ ngrams/buckets off #2891

Merged
merged 6 commits into from
Jul 31, 2020

Conversation

gojomo
Copy link
Collaborator

@gojomo gojomo commented Jul 22, 2020

In looking at the potential for #2879, I'm finding bugs & suboptimalities in FastText. This quickie lets FastText work like Word2Vec, when ngrams are disabled - something that should've worked earlier, but seems to have never been tested.

@gojomo gojomo changed the title [WIP] Ensuring FastText works as well as Word2Vec with ngrams-off Fix FastText word-vectors w/ ngrams/buckets off Jul 23, 2020
@gojomo gojomo changed the title Fix FastText word-vectors w/ ngrams/buckets off [RFM] Fix FastText word-vectors w/ ngrams/buckets off Jul 23, 2020
@gojomo gojomo requested review from piskvorky and mpenkov July 23, 2020 00:22
@gojomo
Copy link
Collaborator Author

gojomo commented Jul 23, 2020

(last commit 8f64b48 doesn't belong here: force-reverting)

gensim/models/fasttext.py Show resolved Hide resolved
gensim/test/test_fasttext.py Outdated Show resolved Hide resolved
gensim/test/test_fasttext.py Outdated Show resolved Hide resolved
@gojomo gojomo changed the title [RFM] Fix FastText word-vectors w/ ngrams/buckets off [MRG] Fix FastText word-vectors w/ ngrams/buckets off Jul 27, 2020
@gojomo
Copy link
Collaborator Author

gojomo commented Jul 31, 2020

Low risk, removes flaw, has test, @piskvorky-reviewed - pushing merge button.

@gojomo gojomo merged commit 779fe46 into piskvorky:develop Jul 31, 2020
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

Successfully merging this pull request may close these issues.

2 participants