Skip to content

fix: cache DataFrames to temp tables in bigframes.bigquery.ml methods to avoid time travel #2318

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
tswast merged 1 commit into from
Dec 9, 2025
Merged

fix: cache DataFrames to temp tables in bigframes.bigquery.ml methods to avoid time travel #2318

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions bigframes/bigquery/_operations/ml.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,6 +39,10 @@ def _to_sql(df_or_sql: Union[pd.DataFrame, dataframe.DataFrame, str]) -> str:
else:
bf_df = cast(dataframe.DataFrame, df_or_sql)

# Cache dataframes to make sure base table is not a snapshot.
# Cached dataframe creates a full copy, never uses snapshot.
# This is a workaround for internal issue b/310266666.
bf_df.cache()
sql, _, _ = bf_df._to_sql_query(include_index=False)
return sql

Expand Down
5 changes: 3 additions & 2 deletions bigframes/ml/core.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -436,8 +436,9 @@ def create_model(
Returns: a BqmlModel, wrapping a trained model in BigQuery
"""
options = dict(options)
# Cache dataframes to make sure base table is not a snapshot
# cached dataframe creates a full copy, never uses snapshot
# Cache dataframes to make sure base table is not a snapshot.
# Cached dataframe creates a full copy, never uses snapshot.
# This is a workaround for internal issue b/310266666.
if y_train is None:
input_data = X_train.reset_index(drop=True).cache()
else:
Expand Down