This repository contains an example notebook providing a deep dive on specializing or improving your Contextual AI agents.
This repository includes the following:
- improvement-overview.ipynb: Jupyter notebook that uses the Python SDK.
Note
This notebook assumes you've completed the Intro End to End Example and already have a Agent setup. If you haven't, please complete that example first.
This notebook has the following steps:
- Queries / Retrievals
- Evaluation Job
- Modifying System Prompt
- Retrieval Settings
- Generation Settings
- Tuning the Agent
The examples can be completed in under 30 minutes. Each notebook is self-contained and includes detailed explanations for each step.
- Contextual AI API Key
- Python 3.8+
- Required dependencies (listed in
requirements.txt)
For comprehensive platform documentation, visit docs.contextual.ai
-
Clone this repository
git clone https://github.com/ContextualAI/examples
-
Install required dependencies
pip install -r requirements.txt
-
Start jupyter lab:
jupyter lab
-
Run through the steps in the notebook
For additional support or questions, please refer to the official documentation or contact the Contextual AI support team.