Skip to content

Adding PUSH from ELI5 Datasets to LangSmith in the Form of CREATE with Splits

Stephan Rayner requested to merge sr/sync-langsmith-push into main

What does this merge request do and why?

This is the other half of syncing data from the ELI5/datasets => LangSmith.

Originally there was a plan to create a fully separate mode under the datasets sync command but it was realised that if we use splits (which are defined in the documentation attached to this MR) we could extend the datasets create command

How to set up and validate locally

  1. Check out to this merge request's branch.
  2. Update your .env file
  3. Install dependencies. 
    poetry install
  4. Check the existing commands ELI5 provides: 
    poetry run eli5 datasets --help
  5. Push a test datasets to LangSmith: 
    poetry run eli5 datasets create <dataset name> <path to the dataset file> --upsert --split <split name>
  6. Wait several minutes and then check in LangSmith

⚠️ After the above make please clean up the dataset on LangSmith, ie delete the split you have created ⚠️

Merge request checklist

  • Tests added for new functionality. If not, please raise an issue to follow up.
  • Documentation added/updated, if needed.

Closes:

Edited by Stephan Rayner

Merge request reports