mirror of
https://github.com/open-webui/open-webui.git
synced 2026-05-07 11:28:35 -05:00
[PR #9290] [CLOSED] automated_ragas allow to automate the evaluation of a set of paramete… #45399
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
📋 Pull Request Information
Original PR: https://github.com/open-webui/open-webui/pull/9290
Author: @brunob31880
Created: 2/3/2025
Status: ❌ Closed
Base:
dev← Head:automated_ragas📝 Commits (1)
06de9aaautomated_ragas allow to automate the evaluation of a set of parameters in openwebui through RAGAS📊 Changes
15 files changed (+1666 additions, -1 deletions)
View changed files
📝
backend/open_webui/config.py(+22 -0)📝
backend/open_webui/env.py(+1 -0)📝
backend/open_webui/main.py(+16 -0)➕
backend/open_webui/ragas/utils.py(+406 -0)➕
backend/open_webui/routers/ragas.py(+146 -0)📝
backend/open_webui/routers/retrieval.py(+61 -1)📝
backend/requirements.txt(+4 -0)➕
src/lib/apis/ragas/index.ts(+194 -0)📝
src/lib/components/admin/Settings.svelte(+25 -0)➕
src/lib/components/admin/Settings/Ragas.svelte(+136 -0)➕
src/lib/components/workspace/Ragas.svelte(+635 -0)📝
src/lib/constants.ts(+1 -0)📝
src/lib/stores/index.ts(+2 -0)📝
src/routes/(app)/workspace/+layout.svelte(+12 -0)➕
src/routes/(app)/workspace/ragas/+page.svelte(+5 -0)📄 Description
Reference to discussion
Ragas Integration #8494 : https://github.com/open-webui/open-webui/discussions/8494
Before submitting, make sure you've checked the following:
Description:
The modification allows you to store the results of an evaluation in a folder, with the aim of later running the RAGAS algorithm on them. You fill in the questions, ground truth model, and documents for your evaluation, then click the button to launch it. This method provides a systematic way to evaluate, using RAGAS, the relevance of a set of parameters.
Note: The section for launching the RAGAS calculation has been commented out. To integrate RAGAS directly into the project, you would need to modify the uvicorn launch command to include the argument --loop asyncio.
Dependency
langchain_openai
And you must fill in the .env your openai key.
🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.