Creating lower dimension embedding of documents

[JoBue64]

Top2Vec runs through until hitting this step and seems to stall/run continuously with Info #276: . omp_set_nested routine deprecated.

[ddangelov]

Can you share the full stack trace?

[JoBue64]

I am not getting an error with a traceback. The routine hangs at last step indicated below. 2025-01-21 15:52:00,013 - top2vec - INFO - Pre-processing documents for training /opt/anaconda3/envs/myenv/lib/python3.10/site-packages/sklearn/feature_extraction/text.py:521 <https://file+.vscode-resource.vscode-cdn.net/opt/anaconda3/envs/myenv/lib/python3.10/site-packages/sklearn/feature_extraction/text.py:521>: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None' warnings.warn( 2025-01-21 15:52:00,613 - top2vec - INFO - Creating vocabulary embedding Embedding vocabulary: 100%|██████████| 13/13 [00:52<00:00, 4.01s/it] 2025-01-21 15:52:53,307 - top2vec - INFO - Create contextualized document embeddings Embedding documents: 100%|██████████| 101/101 [06:49<00:00, 4.05s/it] 3216it [00:00, 11575.21it/s] 2025-01-21 15:59:50,176 - top2vec - INFO - Creating lower dimension embedding of documents OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.

…

Am 21.01.2025 um 15:45 schrieb Dimo Angelov ***@***.***>: Can you share the full stack trace? — Reply to this email directly, view it on GitHub <#367 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BMTHYCUE3Q3OPMS6TUFR7L32LZMP7AVCNFSM6AAAAABVQ6VTX2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMMBUHEZTCMRUGQ>. You are receiving this because you authored the thread.

[JoBue64]

I finally got it to run through, it took cleaning up the virtual environment and reinstalling everything including Python. From the error messages when reinstalling, I believe that gensim was/is the culprit.

…

Am 21.01.2025 um 15:45 schrieb Dimo Angelov ***@***.***>: Can you share the full stack trace? — Reply to this email directly, view it on GitHub <#367 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BMTHYCUE3Q3OPMS6TUFR7L32LZMP7AVCNFSM6AAAAABVQ6VTX2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMMBUHEZTCMRUGQ>. You are receiving this because you authored the thread.