chore: 1.x - nltk upgrade, use `nltk.download('punkt_tab')` - Githubissues

deepset-ai / haystack

:mag: AI orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your data. With advanced retrieval methods, it's best suited for building RAG, question answering, semantic search or conversational agent chatbots.

https://haystack.deepset.ai

Apache License 2.0

16.94k stars 1.85k forks source link

chore: 1.x - nltk upgrade, use `nltk.download('punkt_tab')` #8254

Closed vblagoje closed 1 month ago

vblagoje commented 1 month ago

Pin nltk>=3.9
Instead of nltk.download("punkt") use nltk.download('punkt_tab')
fixes https://github.com/deepset-ai/haystack/issues/8238

vblagoje commented 1 month ago

@julian-risch this one should go on 1.x branch in addition to 1.26.x - is that correct? @anakin87 CI is not being run because base is not 1.x?

anakin87 commented 1 month ago

1.x branch is dead, I think IDK why tests are not running

vblagoje commented 1 month ago

@silvanocerza do you have a suggestion what to do here? More precisely:

Should we add this one to v1.x branch and then cherry-pick to v1.26.x, or just v1.26.x directly?
Any idea why CI is not running on v1.26.x PRs?

vblagoje commented 1 month ago

I got my answer with https://github.com/deepset-ai/haystack/pull/8255 thanks all

vblagoje commented 1 month ago

Closing in favour of https://github.com/deepset-ai/haystack/pull/8256