shapash VS trulens

There's work being done in that regard (like this python module), but as far as I know it's very clearly statistical guesstimates, and though it "works", the mathematical foundations are still somewhat shaky. There are heuristics in there we can't get rid of for now. But it's still better than nothing. Waaaaaay better than nothing.

Shapash – Python library to make machine learning interpretable\ (4 comments)

Try out eurybia, from the author of shapash which is a brilliant library as well.

With the new version of Shapash that is now available, you can document each model you release into production. Within a few lines of code, you can include in an HTML report all the information about your model (and its associated performance), the data it uses, its learning strategy, … this report is designed to be easily shared with a Data Protection Officer, an internal audit department, a risk control department, a compliance department, or anyone who wants to understand his work.

Retrieval using a single vector is called dense passage retrieval (DPR), because an entire passage (dozens to hundreds of tokens) is encoded as a single vector. ColBERT instead encodes a vector-per-token, where each vector is influenced by surrounding context. This leads to meaningfully better results; for example, here’s ColBERT running on Astra DB compared to DPR using openai-v3-small vectors, compared with TruLens for the Braintrust Coda Help Desk data set. ColBERT easily beats DPR at correctness, context relevance, and groundedness.

12. TruLens by TruEra | Github | tutorial

3) Hallucination is probably the biggest problem we solve for. To do evals for hallucination, we typically see our users use a combination of groundedness (does the context support the LLM response) and context relevance (is the retrieved context relevant to the query). There's also a bunch more for the evaluations you mentioned (moderation models, sentiment, usefulness, etc.) and it's pretty easy to add custom evals.

Also - my hot take is that gpt-3.5 is good enough for evals (sometimes better) than gpt-4 if you give the LLM enough instructions on how to do the eval.

website: https://www.trulens.org/

The team at TruEra recently released an open source project for evaluation & tracking of LLM applications called TruLens-Eval. We’ve specifically targeted retrieval-augmented QA as a core use case and so far we’ve seen it used for comparing different models and parameters, prompts, vector-db configurations and query planning strategies. I’d love to get your feedback on it.