Skip to content

A programming framework for agentic AI

fix #493 (#932)

Xiaoyun Zhang4118c8e27f2c
diff --git a/flaml/default/suggest.py b/flaml/default/suggest.py
index 71f4d74d8..9e17687e7 100644
--- a/flaml/default/suggest.py
+++ b/flaml/default/suggest.py
@@ -61,7 +61,8 @@ def load_config_predictor(estimator_name, task, location=None):
return predictor
-def suggest_config(task, X, y, estimator_or_predictor, location=None, k=None):
+def suggest_config(task, X, y, estimator_or_predictor, location=None, k=None, meta_feature_fn=meta_feature):
+
"""Suggest a list of configs for the given task and training data.
The returned configs can be used as starting points for AutoML.fit().
@@ -69,7 +70,7 @@ def suggest_config(task, X, y, estimator_or_predictor, location=None, k=None):
"""
task = (
get_classification_objective(len(np.unique(y)))
- if task == "classification"
+ if task == "classification" and y is not None
else task
)
predictor = (
@@ -86,7 +87,7 @@ def suggest_config(task, X, y, estimator_or_predictor, location=None, k=None):
>= version_parse(older_version)
)
prep = predictor["preprocessing"]
- feature = meta_feature(
+ feature = meta_feature_fn(
task, X_train=X, y_train=y, meta_feature_names=predictor["meta_feature_names"]
)
feature = (np.array(feature) - np.array(prep["center"])) / np.array(prep["scale"])
@@ -99,9 +100,10 @@ def suggest_config(task, X, y, estimator_or_predictor, location=None, k=None):
choice = neighbors[ind]["choice"] if k is None else neighbors[ind]["choice"][:k]
configs = [predictor["portfolio"][x] for x in choice]
for config in configs:
- hyperparams = config["hyperparameters"]
- if hyperparams and "FLAML_sample_size" in hyperparams:
- hyperparams.pop("FLAML_sample_size")
+ if "hyperparameters" in config:
+ hyperparams = config["hyperparameters"]
+ if hyperparams and "FLAML_sample_size" in hyperparams:
+ hyperparams.pop("FLAML_sample_size")
return configs