Deepchecks' SDK

👍

Click here to access ready-to-go templates for common SDK operations.

Deepchecks' Python SDK

Deepchecks LLM Evaluation SDK is a lightweight Python package built on top of Deepchecks LLM Evaluation REST API, to install it simply run pip install deepchecks-llm-client. The SDK allows interacting with the system: uploading the raw interactions, or downloading the interactions enriched with metadata calculated in Deepchecks' Features: properties, topics, annotations, and more.

• Data Upload can be done via batch operations, data streaming, or using langchain tracing.
• Data Download enables retrieving the enriched data from Deepchecks for further analysis, inspection of problematic segments, etc.

📘

For a comprehensive list of all available functionality, please check out the Python SDK Reference

Data Structure: Background

Before diving into the sub-sections describing different logging options we recommend first reading the Hierarchy & Data Upload Format page for a better understanding of the data structure in the Deepchecks system.

Alternatives

📌

Uploading data directly from the UI

Notice - you can also upload and download data to and from the system (csv & xlsx format supported) directly from the UI