Uzay-G
Uzay-G
https://github.com/Uzay-G/espial/blob/main/INSIGHTS.md#algorithm-issues
Note: we might have a clustering problem where certain files are over-represented: how to fix to optimize good results?
Yeah this would be nice, currently failing on release candidates for example (vX.YrcN), the format proposed by PyPi.
I also have this problem.
Did anyone manage to get this working?
See https://github.com/ddangelov/Top2Vec/issues/236#issuecomment-1115927332
I get this also with universal sentence decoder when I import top2vec. I think it's because on python 3.10, numba requires numpy =1.22 (https://github.com/RaRe-Technologies/gensim/blob/93ed2fbe27dab41b459bd09bd5b19ed5d7e01908/.github/workflows/build-wheels.yml#L95). So the library is not usable...
How to fix this: use python < 3.10, use numpy
The problem is with python 3.10
3.10 is >= 3.6? I don't mean 3.1, python 3.10 (https://www.python.org/downloads/release/python-3100/) was released in october.