Julian Risch

Alright, good luck with your thesis! 🎓

@sjrl Yes, it's about porting TransformersQueryClassifier but could become two classes TransformersTextClassifier and TransformersDocumentClassifier. That would be consistent with the SentenceTransformersDocumentEmbedder and the SentenceTransformersTextEmbedder we have in 2.x. Should behave...

@so2liu Thank you for bringing up this feature request. Let's see whether somebody in our open source community with expertise in Chinese language wants to contribute here. the line of...

`stop_words` are so closely related to `PromptTemplate` that I suggest that `stop_words` could become part of `PromptTemplate`. An optional parameter in `PromptTemplate`. I linked this issue to the Prompt Hub...

@rnyak Thank you for this feature suggestion. I completely agree that the parameters should be more similar for DPR and EmbeddingRetriever training to make the training easier to use. 👍...

There is the document and the document offset. I would indeed suggest to cut out the context and set the context offset then. Can be a separate component that gets...

I see here that the 1.x reader added the document page number to the meta field of the Answer based on the document offset. We could add the same functionality...

@vineetsajuTR We'll work on this topic this quarter.

@sjrl We are considering this issue for our next sprint. Is there any new info that will be relevant for the implementation of this component?

@davidsbatista Could you please check again if we can merge the two issues https://github.com/deepset-ai/haystack/issues/5700 and https://github.com/deepset-ai/haystack/issues/5702 or whether they should remain separate?