chore: fix 'multimodel' typo to 'multimodal' (#17038)

Corrects the typo "multimodel" to "multimodal" across the codebase. This
includes internal method names, comments, test functions, and GCS paths
in system tests. This aligns the naming with the standard term for
models supporting multiple data modalities.

Fixes #<issue_number_goes_here> 🦕

Author: shuoweil

packages/bigframes/bigframes/bigquery/_operations/ai.py

@@ -1205,7 +1205,7 @@ def _convert_series(
     result = convert.to_bf_series(s, default_index=None, session=session)
 
     if result.dtype == dtypes.OBJ_REF_DTYPE:
-        # Support multimodel
+        # Support multimodal
         return result.blob.read_url()
     return result
 

packages/bigframes/bigframes/operations/ai.py

@@ -223,7 +223,7 @@ def map(
                 bigframes.series.Series,
                 model.predict(
                     df,
-                    prompt=self._make_multimodel_prompt(
+                    prompt=self._make_multimodal_prompt(
                         df, columns, user_instruction, output_instruction
                     ),
                     temperature=0.0,
@@ -774,7 +774,7 @@ def _attach_embedding(dataframe, source_column: str, embedding_column: str, mode
         return result_df
 
     @staticmethod
-    def _make_multimodel_prompt(
+    def _make_multimodal_prompt(
         prompt_df, columns, user_instruction: str, output_instruction: str
     ):
         prompt = [f"{output_instruction}\n{user_instruction}\nContext: "]

packages/bigframes/bigframes/operations/semantics.py

@@ -397,7 +397,7 @@ def filter(self, instruction: str, model, ground_with_google_search: bool = Fals
                 bigframes.dataframe.DataFrame,
                 model.predict(
                     df,
-                    prompt=self._make_multimodel_prompt(
+                    prompt=self._make_multimodal_prompt(
                         df, columns, user_instruction, output_instruction
                     ),
                     temperature=0.0,
@@ -518,7 +518,7 @@ def map(
                 bigframes.series.Series,
                 model.predict(
                     df,
-                    prompt=self._make_multimodel_prompt(
+                    prompt=self._make_multimodal_prompt(
                         df, columns, user_instruction, output_instruction
                     ),
                     temperature=0.0,
@@ -1094,7 +1094,7 @@ def _attach_embedding(dataframe, source_column: str, embedding_column: str, mode
         return result_df
 
     @staticmethod
-    def _make_multimodel_prompt(
+    def _make_multimodal_prompt(
         prompt_df, columns, user_instruction: str, output_instruction: str
     ):
         prompt = [f"{output_instruction}\n{user_instruction}\nContext: "]

packages/bigframes/tests/system/large/operations/test_ai.py

@@ -65,7 +65,7 @@ def test_filter_multi_model(session, gemini_flash_model):
         10,
     ):
         df = session.from_glob_path(
-            "gs://bigframes-dev-testing/a_multimodel/images/*", name="image"
+            "gs://bigframes-dev-testing/a_multimodal/images/*", name="image"
         )
         df["prey"] = series.Series(
             ["building", "cross road", "rock", "squirrel", "rabbit"], session=session
@@ -229,7 +229,7 @@ def test_map(session, gemini_flash_model, output_schema, output_col):
     )
 
 
-def test_map_multimodel(session, gemini_flash_model):
+def test_map_multimodal(session, gemini_flash_model):
     with bigframes.option_context(
         AI_OP_EXP_OPTION,
         True,
@@ -239,7 +239,7 @@ def test_map_multimodel(session, gemini_flash_model):
         10,
     ):
         df = session.from_glob_path(
-            "gs://bigframes-dev-testing/a_multimodel/images/*", name="image"
+            "gs://bigframes-dev-testing/a_multimodal/images/*", name="image"
         )
         df["scenario"] = series.Series(
             ["building", "cross road", "tree", "squirrel", "rabbit"], session=session

packages/bigframes/tests/system/large/operations/test_semantics.py

@@ -412,7 +412,7 @@ def test_filter_multi_model(session, gemini_flash_model):
         10,
     ):
         df = session.from_glob_path(
-            "gs://bigframes-dev-testing/a_multimodel/images/*", name="image"
+            "gs://bigframes-dev-testing/a_multimodal/images/*", name="image"
         )
         df["prey"] = series.Series(
             ["building", "cross road", "rock", "squirrel", "rabbit"], session=session
@@ -571,7 +571,7 @@ def test_map(session, gemini_flash_model):
     )
 
 
-def test_map_multimodel(session, gemini_flash_model):
+def test_map_multimodal(session, gemini_flash_model):
     with bigframes.option_context(
         SEM_OP_EXP_OPTION,
         True,
@@ -581,7 +581,7 @@ def test_map_multimodel(session, gemini_flash_model):
         10,
     ):
         df = session.from_glob_path(
-            "gs://bigframes-dev-testing/a_multimodel/images/*", name="image"
+            "gs://bigframes-dev-testing/a_multimodal/images/*", name="image"
         )
         df["scenario"] = series.Series(
             ["building", "cross road", "tree", "squirrel", "rabbit"], session=session

packages/bigframes/tests/system/small/bigquery/test_ai.py

@@ -160,7 +160,7 @@ def test_ai_generate_bool(session):
 
 def test_ai_generate_bool_multi_model(session):
     df = session.from_glob_path(
-        "gs://bigframes-dev-testing/a_multimodel/images/*", name="image"
+        "gs://bigframes-dev-testing/a_multimodal/images/*", name="image"
     )
 
     result = bbq.ai.generate_bool((df["image"], " contains an animal"))
@@ -197,7 +197,7 @@ def test_ai_generate_int(session):
 
 def test_ai_generate_int_multi_model(session):
     df = session.from_glob_path(
-        "gs://bigframes-dev-testing/a_multimodel/images/*", name="image"
+        "gs://bigframes-dev-testing/a_multimodal/images/*", name="image"
     )
 
     result = bbq.ai.generate_int(
@@ -236,7 +236,7 @@ def test_ai_generate_double(session):
 
 def test_ai_generate_double_multi_model(session):
     df = session.from_glob_path(
-        "gs://bigframes-dev-testing/a_multimodel/images/*", name="image"
+        "gs://bigframes-dev-testing/a_multimodal/images/*", name="image"
     )
 
     result = bbq.ai.generate_double(
@@ -307,7 +307,7 @@ def test_ai_if(session):
 
 def test_ai_if_multi_model(session, bq_connection):
     df = session.from_glob_path(
-        "gs://bigframes-dev-testing/a_multimodel/images/*",
+        "gs://bigframes-dev-testing/a_multimodal/images/*",
         name="image",
         connection=bq_connection,
     )
@@ -338,7 +338,7 @@ def test_ai_classify_with_examples(session):
 
 def test_ai_classify_multi_model(session, bq_connection):
     df = session.from_glob_path(
-        "gs://bigframes-dev-testing/a_multimodel/images/*",
+        "gs://bigframes-dev-testing/a_multimodal/images/*",
         name="image",
         connection=bq_connection,
     )
@@ -361,7 +361,7 @@ def test_ai_score(session):
 
 def test_ai_score_multi_model(session):
     df = session.from_glob_path(
-        "gs://bigframes-dev-testing/a_multimodel/images/*", name="image"
+        "gs://bigframes-dev-testing/a_multimodal/images/*", name="image"
     )
     prompt = ("Rank the liveliness of ", df["image"], "on the scale from 1 to 3")