How do I delete score columns creating when evaluating datasets via the API?

Is there a way for me to delete these columns from either the UI or the API?

Solution:

Makes sense, datasets right now use the full history to render score columns. Agree that only using the traces that are actually in the dataset makes more sense here. Can you add this as a bug? https://langfuse.com/issues

GitHub

Issues · langfuse/langfuse

🪢 Open source LLM engineering platform: LLM Observability, metrics, evals, prompt management, playground, datasets. Integrates with LlamaIndex, Langchain, OpenAI SDK, LiteLLM, and more. 🍊YC W23 - ...

Jump to solution

3 Replies

Marc•2d ago

Only scores used during the timeframe that you select on the table will be shown, thus if you stop using these eval scores, they will stop to show up in the table. currently they cannot be deleted manually

micah.stairs•2d ago

Thanks for the speedy response! I'm using v2.76.0 and I don't see an option in the dataruns table to filter by a timeframe. Also I deleted the entire dataset where these obsolete columns were used

Solution

Marc•this hour

GitHub

Issues · langfuse/langfuse