Skip to content

A programming framework for agentic AI

Update Tune-User-Defined-Function.md (#562)

Chi Wang7748e0ff49b3
diff --git a/website/docs/Use-Cases/Tune-User-Defined-Function.md b/website/docs/Use-Cases/Tune-User-Defined-Function.md
index 6f804b131..b46df63b8 100644
--- a/website/docs/Use-Cases/Tune-User-Defined-Function.md
+++ b/website/docs/Use-Cases/Tune-User-Defined-Function.md
@@ -4,7 +4,8 @@
1. Your machine learning task is not one of the built-in tasks from `flaml.AutoML`.
1. Your input cannot be represented as X_train + y_train or dataframe + label.
-1. You want to tune a function that may not even be a machine learning procedure.
+1. The optimization metric is not measurable via validation data only. For example, when you want to directly optimize a downstream application instead of a model accuracy metric.
+1. You need to tune a function that may not even be a machine learning procedure.
## Basic Tuning Procedure