Julian Risch
Julian Risch
Instead of jumping right into implementation, let's take the time to collect requirements and design a solution that addresses them. Related issue about streaming: https://github.com/deepset-ai/haystack/issues/4890 Related issue about deployment guides:...
**Describe the bug** Although I set the `stopwords` parameter of `HuggingFaceLocalGenerator` to `["Original"]` it keeps on generating after this token was generated. The only effect of setting the stopword is...
### Related Issues - fixes #7570 ### Proposed Changes: - Calculate document page number of ExtractedAnswer as in Haystack 1.x - Add page number to ExtractedAnswer's meta data under the...
When users switch between statistical evaluators and model-based evaluators the input and output names of the component often change. We should make them more consistent. **Output Names** For example, every...
Currently, preprocessing components such as the DocumentCleaner get documents as input and return preprocessed documents **but with different ids**. We should enable users to specify custom ids and make sure...
Goal: Showcase the Haystack evaluation metrics on an example that is close to what our users are trying to do, i.e. to improve a retriever in a RAG app. The...
This issue is for tracking the implementation of the proposal https://github.com/deepset-ai/haystack/pull/6784 by creating integrations for [Uptrain](https://docs.uptrain.ai/getting-started/introduction), [RAGAS](https://docs.ragas.io/en/stable/index.html) and [DeepEval](https://docs.confident-ai.com/docs/getting-started), as well as a custom implementation integrated in Haystsack. ```[tasklist] ###...
We need to extend evaluation features, in particular for RAG pipelines so that users can answer questions like: - Is this pipeline good enough? - What should I focus on...
- [ ] Colab code works as expected - [ ] Colab has view access for the general public - [ ] Colab has been added to cookbook repository -...