|
26 | 26 | from bigframes import clients, dataframe, dtypes |
27 | 27 | from bigframes import pandas as bpd |
28 | 28 | from bigframes import series, session |
29 | | -from bigframes.bigquery._operations import utils as ml_utils |
| 29 | +from bigframes.bigquery._operations import utils as bq_utils |
30 | 30 | from bigframes.core import convert |
31 | 31 | from bigframes.core.logging import log_adapter |
32 | 32 | import bigframes.core.sql.literals |
@@ -455,8 +455,8 @@ def generate_embedding( |
455 | 455 | for details. |
456 | 456 | """ |
457 | 457 | data = _to_dataframe(data, series_rename="content") |
458 | | - model_name, session = ml_utils.get_model_name_and_session(model, data) |
459 | | - table_sql = ml_utils.to_sql(data) |
| 458 | + model_name, session = bq_utils.get_model_name_and_session(model, data) |
| 459 | + table_sql = bq_utils.to_sql(data) |
460 | 460 |
|
461 | 461 | struct_fields: Dict[str, bigframes.core.sql.literals.STRUCT_VALUES] = {} |
462 | 462 | if output_dimensionality is not None: |
@@ -564,8 +564,8 @@ def generate_text( |
564 | 564 | The generated text. |
565 | 565 | """ |
566 | 566 | data = _to_dataframe(data, series_rename="prompt") |
567 | | - model_name, session = ml_utils.get_model_name_and_session(model, data) |
568 | | - table_sql = ml_utils.to_sql(data) |
| 567 | + model_name, session = bq_utils.get_model_name_and_session(model, data) |
| 568 | + table_sql = bq_utils.to_sql(data) |
569 | 569 |
|
570 | 570 | struct_fields: Dict[ |
571 | 571 | str, |
|
0 commit comments