This example notebook, "Customer Support Classification.ipynb," demonstrates setting up a comprehensive workflow for classifying customer support requests. It walks through the setup of a project that integrates with the LLM Interactive ML Backend example to provide an LLM prompter as a predictor in the Label Studio UI.
To use this notebook:
- Clone this repository to your local machine.
- Start Label Studio
- Start the LLM Interactive ML Backend example (Make sure to configure your OpenAI API key)
- Run the "Customer Support Classification.ipynb" notebook in Jupyter.
- Connect the created project to the ML Backend as described here. Make sure to enable "Use for interactive preannotations."