LLM Evaluation in Haystack

[julian-risch]

This issue is for tracking the implementation of the proposal #6784 by creating integrations for Uptrain, RAGAS and DeepEval, as well as a custom implementation integrated in Haystsack.

Tasks

Preview Give feedback

1. Add UpTrain evaluation framework integration haystack-core-integrations#248
   new integration +1
   
2. Add RAGAS evaluation framework integration haystack-core-integrations#249
   P1 integration:ragas new integration +3
   
3. Add DeepEval evaluation framework integration haystack-core-integrations#250
   P1 new integration +2
   
4. Write a documentation page with an overview of model-based eval frameworks #6788
   type:documentation +1
   
5. Create a tutorial about model-based evaluation of a RAG pipeline #6789
   +0
   
6. Write documentation page about evaluation in general #6787
   type:documentation +1
   
7. Update tutorial on model-based evaluation with Haystack #6790
   2.x P2 topic:eval +3
   
8. Write a blogpost to announce LLM evaluation support in Haystack #6791
   2.x topic:eval +2
   
9. Custom LLM-based evaluator in Haystack core #7022
   3 of 3
   2.x topic:eval +2
   
10. Design output API of evaluation #7415
    2 of 2
    P1 topic:eval +2
    
11. Showcase Haystack evaluations on an industry dataset #7407
    2 of 2
    P2 epic topic:eval +3
12. Cookbook for Ragas haystack-cookbook#52
    2.x Contributions wanted! P1 +3
    
13. Cookbook for DeepEval haystack-cookbook#53
    2.x Contributions wanted! P1 +3
    
14. docs: Add documentation pages for new evaluation components #7471
    2.x P1 epic topic:eval type:documentation +5