Stefano Fiorucci
Stefano Fiorucci
As stated in #2965, it would be good to include in Tutorial 14 a small section about new features of `TransformersQueryClassifier`: - using custom transformer models - using zero-shot-classification ###...
**Describe the bug** When passed a list of files with different extensions, `FileTypeClassifier` throws an error. Reading the code, this behavior is expected, but I can't understand why... https://github.com/deepset-ai/haystack/blob/c91316e862c3fb751b3e8996ddd5f99b5563ae81/haystack/nodes/file_classifier/file_type.py#L74-L78 **Shouldn't...
### Related Issues - fixes #2589 ### Proposed Changes: **First draft** to manually temporarily fix this problem. When Pydantic v2 will be released, there will be no need of this...
**Describe the bug** Importing audio nodes (such as `haystack.nodes.audio.answer_to_speech`) raises an error **Error message** > Traceback (most recent call last): > File "__init__.pxd", line 1035, in numpy.import_array > RuntimeError: module...
I installed ghostscript by the command "apt install ghostscript". I noticed that, using ghostscript 9.26, Camelot can't detect various tables and it doesn't show any error message. For example, using...
I tried to extract tables from this [PDF](https://github.com/camelot-dev/camelot/files/3950873/191204A61SO2.pdf) with the following code: `tables=camelot.read_pdf('mydoc.pdf', flavor='lattice' , line_scale=80, pages='1-end')` **Error** Traceback (most recent call last): File "/usr/src/app/swagger_server/services/extract_tables_from_pdf.py", line 108, in extract_tables_from_path tables...
### Related Issues - fixes #3153 ### Proposed Changes: `batch_size` parameter wasn't used in `get_documents_by_id`. Now the method uses `batch_size`, making several queries based on this parameter. Implementation inspired by...
**Describe the bug** Reported on Discord: using `OpenAIAnswerGenerator` with gpt-3.5-turbo raises an error. *Generator vs PromptNode* The `OpenAIAnswerGenerator` is still available in Haystack while the `PromptNode` is the preferred choice....
Pydantic V2 was recently released. The new version was not compatible with Haystack: [here you can see the errors](https://github.com/deepset-ai/haystack/actions/runs/5425007074/jobs/9865473307) that lead to pinning `pydantic
Related issue: https://github.com/deepset-ai/haystack-tutorials/issues/220 **Plan**: - Retrain the models with `scikit-learn==1.3.0` and add the models to publish in this issue (they should be published when Haystack 1.19.0 is released, to prevent...