How to: Evals

This section contains practical guides for working with evaluators in Phoenix.

• Custom LLM Evaluators - Build custom classification and scoring evaluators

• Code Evaluators - Create non-LLM evaluators using functions

• Batch Evaluations - Run evaluations on dataframes efficiently

• Using Evals with Phoenix - Integrate evaluators with traces and experiments

• Configuring the LLM - Configure LLM models for evaluators