search

GoogleCloudPlatform / applied-ai-engineering-samples

This repository compiles code samples and notebooks demonstrating how to use Generative AI on Google Cloud Vertex AI.
https://googlecloudplatform.github.io/applied-ai-engineering-samples/
Apache License 2.0
357 stars 129 forks source link

added rag eval dataset curation notebook using gemini for cluster ana… #108

Closed kenleejr closed 1 week ago

kenleejr commented 1 month ago

The purpose of this notebook is to go from a raw a set of user-queries from an in-production chatbot or rag system to a refined evaluation dataset. This is achieved through a mix of traditional clustering and using gemini to generate structured outputs from the unstructured chat data.

  1. Embed questions
  2. Cluster using favorite clustering algorithm
  3. Summarize each cluster using gemini, extracting valuable metadata like sentiment, topics, representative questions, etc.

Best,

review-notebook-app[bot] commented 1 month ago

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

kenleejr commented 1 month ago

Hi @Abhishekbhagwat I've fixed all the changes you noted, let me know if you would like more.

RajeshThallam commented 2 weeks ago

@kenleejr Can you please add a README with brief description to your folder as well?