microsoft / microsoft/rag-experiment-accelerator

Separate each search type into its own mlflow run to allow comparison using Azure ML / mlflow

Open
#540 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
311
Forks
111
PR merge metrics
No merged PRs in 30d

Description

AS a researcher I would like to be able to compare metric results across different types of search approaches in Azure ML / ML Flow
SO, I would be able to choose best type of search for my case.

related to #529

Note: currently eval step averages all metrics for all search types and log to mflow only the mean value per metric, but also uploads the full detailed table to azureml.

DoD:

  • query and eval steps will run per search type in parallel (like Index does with IndexConfig class)
### Tasks

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Trace the query and eval steps and the existing parallel execution based on the IndexConfig class first. Done means query and eval run separately for each search type in parallel, with metrics logged per type so results can be compared rather than only averaged.

Written by the indexing model from the issue text.

Assessment

Tech stack
azure, python
Domain
data, machine-learning, search
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.