False positive for PD011 #14669
Description
I am getting this:
src/utils/embeddings.py:47:17: PD011 Use `.to_numpy()` instead of `.values`
|
45 | )
46 | logger.info("Successfully generated embeddings")
47 | return [embedding.values for embedding in embeddings]
| ^^^^^^^^^^^^^^^^ PD011
48 | except ValueError as e:
49 | logger.error("Failed to get embeddings due to an API error: %s", e)
But the type embedding does not have a .to_numpy method.
Here is the embedding class:
@dataclasses.dataclass
class TextEmbedding:
"""Text embedding vector and statistics."""
__module__ = "vertexai.language_models"
values: List[float]
statistics: Optional[TextEmbeddingStatistics] = None
_prediction_response: Optional[aiplatform.models.Prediction] = None
@classmethod
def _parse_text_embedding_response(
cls, prediction_response: aiplatform.models.Prediction, prediction_index: int
) -> "TextEmbedding":
"""Creates a `TextEmbedding` object from a prediction.
Args:
prediction_response: `aiplatform.models.Prediction` object.
Returns:
`TextEmbedding` object.
"""
prediction = prediction_response.predictions[prediction_index]
is_prediction_from_pretrained_models = isinstance(
prediction, collections.abc.Mapping
)
if is_prediction_from_pretrained_models:
embeddings = prediction["embeddings"]
embedding_stats = embeddings["statistics"]
return cls(
values=embeddings["values"],
statistics=TextEmbeddingStatistics(
token_count=embedding_stats["token_count"],
truncated=embedding_stats["truncated"],
),
_prediction_response=prediction_response,
)
else:
return cls(values=prediction, _prediction_response=prediction_response)This code is derived from .venv/lib/python3.12/site-packages/vertexai/language_models/_language_models.py,