Skip to content

A programming framework for agentic AI

fix NLP zero division error (#1009)

Susan Xueqing Liu00c30a398e6b
diff --git a/flaml/automl/model.py b/flaml/automl/model.py
index a04f90442..c32a89f35 100644
--- a/flaml/automl/model.py
+++ b/flaml/automl/model.py
@@ -1135,7 +1135,7 @@ class TransformersEstimator(BaseEstimator):
predictions = new_trainer.predict(test_dataset).predictions
except ZeroDivisionError:
logger.warning("Zero division error appeared in HuggingFace Transformers.")
- predictions = np.array([-0.05] * len(test_dataset))
+ predictions = None
return predictions
def score(self, X_val: DataFrame, y_val: Series, **kwargs):
@@ -1171,7 +1171,7 @@ class TransformersEstimator(BaseEstimator):
predictions = new_trainer.predict(test_dataset, **kwargs).predictions
except ZeroDivisionError:
logger.warning("Zero division error appeared in HuggingFace Transformers.")
- predictions = np.array([0] * len(test_dataset))
+ predictions = None
post_y_pred, _ = postprocess_prediction_and_true(
task=self._task,
y_pred=predictions,
diff --git a/flaml/automl/nlp/huggingface/utils.py b/flaml/automl/nlp/huggingface/utils.py
index 88a555578..9f07be368 100644
--- a/flaml/automl/nlp/huggingface/utils.py
+++ b/flaml/automl/nlp/huggingface/utils.py
@@ -311,6 +311,8 @@ def tokenize_swag(this_row, tokenizer, hf_args=None, return_column_name=False):
def postprocess_prediction_and_true(task, y_pred, tokenizer, hf_args, y_true=None, X=None):
# postprocess the matrix prediction y_pred and ground truth y_true into user readable format, e.g., for summarization, decode into text
+ if y_pred is None:
+ return np.array([0.0] * len(X)), y_true
if task == SEQCLASSIFICATION:
return np.argmax(y_pred, axis=1), y_true
elif task == SEQREGRESSION: