feat: Adds additional_experiments field to AutoMlTablesInputs (#544)

PiperOrigin-RevId: 384715948

Source-Link: googleapis/googleapis@83c3728

Source-Link: googleapis/googleapis-gen@4c0c40d

feat: Adds additional_experiments field to AutoMlForecastingInputs 
feat: Adds two new ModelType constants for Video Action Recognition training jobs

google/cloud/aiplatform/v1beta1/schema/predict/instance_v1beta1/types/image_classification.py

@@ -26,8 +26,8 @@ class ImageClassificationPredictionInstance(proto.Message):
     r"""Prediction input format for Image Classification.
     Attributes:
         content (str):
-            The image bytes or GCS URI to make the
-            prediction on.
+            The image bytes or Cloud Storage URI to make
+            the prediction on.
         mime_type (str):
             The MIME type of the content of the image.
             Only the images in below listed MIME types are

google/cloud/aiplatform/v1beta1/schema/predict/instance_v1beta1/types/image_object_detection.py

@@ -26,8 +26,8 @@ class ImageObjectDetectionPredictionInstance(proto.Message):
     r"""Prediction input format for Image Object Detection.
     Attributes:
         content (str):
-            The image bytes or GCS URI to make the
-            prediction on.
+            The image bytes or Cloud Storage URI to make
+            the prediction on.
         mime_type (str):
             The MIME type of the content of the image.
             Only the images in below listed MIME types are

google/cloud/aiplatform/v1beta1/schema/predict/instance_v1beta1/types/text_extraction.py

@@ -36,7 +36,7 @@ class TextExtractionPredictionInstance(proto.Message):
             If a key is provided, the batch prediction
             result will by mapped to this key. If omitted,
             then the batch prediction result will contain
-            the entire input instance. AI Platform will not
+            the entire input instance. Vertex AI will not
             check if keys in the request are duplicates, so
             it is up to the caller to ensure the keys are
             unique.

google/cloud/aiplatform/v1beta1/schema/predict/params_v1beta1/types/video_classification.py

@@ -37,16 +37,16 @@ class VideoClassificationPredictionParams(proto.Message):
             10,000.
         segment_classification (bool):
             Set to true to request segment-level
-            classification. AI Platform returns labels and
+            classification. Vertex AI returns labels and
             their confidence scores for the entire time
             segment of the video that user specified in the
             input instance. Default value is true
         shot_classification (bool):
             Set to true to request shot-level
-            classification. AI Platform determines the
+            classification. Vertex AI determines the
             boundaries for each camera shot in the entire
             time segment of the video that user specified in
-            the input instance. AI Platform then returns
+            the input instance. Vertex AI then returns
             labels and their confidence scores for each
             detected shot, along with the start and end time
             of the shot.
@@ -57,15 +57,14 @@ class VideoClassificationPredictionParams(proto.Message):
             Default value is false
         one_sec_interval_classification (bool):
             Set to true to request classification for a
-            video at one-second intervals. AI Platform
-            returns labels and their confidence scores for
-            each second of the entire time segment of the
-            video that user specified in the input WARNING:
-            Model evaluation is not done for this
-            classification type, the quality of it depends
-            on the training data, but there are no metrics
-            provided to describe that quality. Default value
-            is false
+            video at one-second intervals. Vertex AI returns
+            labels and their confidence scores for each
+            second of the entire time segment of the video
+            that user specified in the input WARNING: Model
+            evaluation is not done for this classification
+            type, the quality of it depends on the training
+            data, but there are no metrics provided to
+            describe that quality. Default value is false
     """
 
     confidence_threshold = proto.Field(proto.FLOAT, number=1,)

google/cloud/aiplatform/v1beta1/schema/predict/prediction_v1beta1/types/classification.py

@@ -27,8 +27,7 @@ class ClassificationPredictionResult(proto.Message):
     Attributes:
         ids (Sequence[int]):
             The resource IDs of the AnnotationSpecs that
-            had been identified, ordered by the confidence
-            score descendingly.
+            had been identified.
         display_names (Sequence[str]):
             The display names of the AnnotationSpecs that
             had been identified, order matches the IDs.

google/cloud/aiplatform/v1beta1/schema/trainingjob/definition_v1beta1/types/automl_tables.py

@@ -139,6 +139,9 @@ class AutoMlTablesInputs(proto.Message):
             predictions to a BigQuery table. If this
             configuration is absent, then the export is not
             performed.
+        additional_experiments (Sequence[str]):
+            Additional experiment flags for the Tables
+            training pipeline.
     """
 
     class Transformation(proto.Message):
@@ -401,6 +404,7 @@ class TextArrayTransformation(proto.Message):
         number=10,
         message=gcastd_export_evaluated_data_items_config.ExportEvaluatedDataItemsConfig,
     )
+    additional_experiments = proto.RepeatedField(proto.STRING, number=11,)
 
 
 class AutoMlTablesMetadata(proto.Message):

google/cloud/aiplatform/v1beta1/schema/trainingjob/definition_v1beta1/types/automl_time_series_forecasting.py

@@ -167,6 +167,9 @@ class AutoMlForecastingInputs(proto.Message):
 
             -  "ignore-validation" - ignore the results of the
                validation and continue
+        additional_experiments (Sequence[str]):
+            Additional experiment flags for the time
+            series forcasting training.
     """
 
     class Transformation(proto.Message):
@@ -376,6 +379,7 @@ class Granularity(proto.Message):
     )
     quantiles = proto.RepeatedField(proto.DOUBLE, number=16,)
     validation_options = proto.Field(proto.STRING, number=17,)
+    additional_experiments = proto.RepeatedField(proto.STRING, number=25,)
 
 
 class AutoMlForecastingMetadata(proto.Message):

google/cloud/aiplatform/v1beta1/schema/trainingjob/definition_v1beta1/types/automl_video_action_recognition.py

@@ -48,6 +48,8 @@ class ModelType(proto.Enum):
         MODEL_TYPE_UNSPECIFIED = 0
         CLOUD = 1
         MOBILE_VERSATILE_1 = 2
+        MOBILE_JETSON_VERSATILE_1 = 3
+        MOBILE_CORAL_VERSATILE_1 = 4
 
     model_type = proto.Field(proto.ENUM, number=1, enum=ModelType,)
 

google/cloud/aiplatform/v1beta1/schema/trainingjob/definition_v1beta1/types/export_evaluated_data_items_config.py

@@ -33,7 +33,6 @@ class ExportEvaluatedDataItemsConfig(proto.Message):
 
             If not specified, then results are exported to the following
             auto-created BigQuery table:
-
             <project_id>:export_evaluated_examples_<model_name>_<yyyy_MM_dd'T'HH_mm_ss_SSS'Z'>.evaluated_examples
         override_existing_table (bool):
             If true and an export destination is

google/cloud/aiplatform_v1/services/migration_service/client.py

@@ -195,32 +195,32 @@ def parse_dataset_path(path: str) -> Dict[str, str]:
         return m.groupdict() if m else {}
 
     @staticmethod
-    def dataset_path(project: str, dataset: str,) -> str:
+    def dataset_path(project: str, location: str, dataset: str,) -> str:
         """Returns a fully-qualified dataset string."""
-        return "projects/{project}/datasets/{dataset}".format(
-            project=project, dataset=dataset,
+        return "projects/{project}/locations/{location}/datasets/{dataset}".format(
+            project=project, location=location, dataset=dataset,
         )
 
     @staticmethod
     def parse_dataset_path(path: str) -> Dict[str, str]:
         """Parses a dataset path into its component segments."""
-        m = re.match(r"^projects/(?P<project>.+?)/datasets/(?P<dataset>.+?)$", path)
+        m = re.match(
+            r"^projects/(?P<project>.+?)/locations/(?P<location>.+?)/datasets/(?P<dataset>.+?)$",
+            path,
+        )
         return m.groupdict() if m else {}
 
     @staticmethod
-    def dataset_path(project: str, location: str, dataset: str,) -> str:
+    def dataset_path(project: str, dataset: str,) -> str:
         """Returns a fully-qualified dataset string."""
-        return "projects/{project}/locations/{location}/datasets/{dataset}".format(
-            project=project, location=location, dataset=dataset,
+        return "projects/{project}/datasets/{dataset}".format(
+            project=project, dataset=dataset,
         )
 
     @staticmethod
     def parse_dataset_path(path: str) -> Dict[str, str]:
         """Parses a dataset path into its component segments."""
-        m = re.match(
-            r"^projects/(?P<project>.+?)/locations/(?P<location>.+?)/datasets/(?P<dataset>.+?)$",
-            path,
-        )
+        m = re.match(r"^projects/(?P<project>.+?)/datasets/(?P<dataset>.+?)$", path)
         return m.groupdict() if m else {}
 
     @staticmethod

google/cloud/aiplatform_v1beta1/services/migration_service/client.py

@@ -179,19 +179,16 @@ def parse_annotated_dataset_path(path: str) -> Dict[str, str]:
         return m.groupdict() if m else {}
 
     @staticmethod
-    def dataset_path(project: str, location: str, dataset: str,) -> str:
+    def dataset_path(project: str, dataset: str,) -> str:
         """Returns a fully-qualified dataset string."""
-        return "projects/{project}/locations/{location}/datasets/{dataset}".format(
-            project=project, location=location, dataset=dataset,
+        return "projects/{project}/datasets/{dataset}".format(
+            project=project, dataset=dataset,
         )
 
     @staticmethod
     def parse_dataset_path(path: str) -> Dict[str, str]:
         """Parses a dataset path into its component segments."""
-        m = re.match(
-            r"^projects/(?P<project>.+?)/locations/(?P<location>.+?)/datasets/(?P<dataset>.+?)$",
-            path,
-        )
+        m = re.match(r"^projects/(?P<project>.+?)/datasets/(?P<dataset>.+?)$", path)
         return m.groupdict() if m else {}
 
     @staticmethod
@@ -211,16 +208,19 @@ def parse_dataset_path(path: str) -> Dict[str, str]:
         return m.groupdict() if m else {}
 
     @staticmethod
-    def dataset_path(project: str, dataset: str,) -> str:
+    def dataset_path(project: str, location: str, dataset: str,) -> str:
         """Returns a fully-qualified dataset string."""
-        return "projects/{project}/datasets/{dataset}".format(
-            project=project, dataset=dataset,
+        return "projects/{project}/locations/{location}/datasets/{dataset}".format(
+            project=project, location=location, dataset=dataset,
         )
 
     @staticmethod
     def parse_dataset_path(path: str) -> Dict[str, str]:
         """Parses a dataset path into its component segments."""
-        m = re.match(r"^projects/(?P<project>.+?)/datasets/(?P<dataset>.+?)$", path)
+        m = re.match(
+            r"^projects/(?P<project>.+?)/locations/(?P<location>.+?)/datasets/(?P<dataset>.+?)$",
+            path,
+        )
         return m.groupdict() if m else {}
 
     @staticmethod

schema/predict/instance/google/cloud/aiplatform/v1beta1/schema/predict/instance_v1beta1/types/image_classification.py

@@ -28,8 +28,8 @@ class ImageClassificationPredictionInstance(proto.Message):
     r"""Prediction input format for Image Classification.
     Attributes:
         content (str):
-            The image bytes or GCS URI to make the
-            prediction on.
+            The image bytes or Cloud Storage URI to make
+            the prediction on.
         mime_type (str):
             The MIME type of the content of the image.
             Only the images in below listed MIME types are

schema/predict/instance/google/cloud/aiplatform/v1beta1/schema/predict/instance_v1beta1/types/image_object_detection.py

@@ -28,8 +28,8 @@ class ImageObjectDetectionPredictionInstance(proto.Message):
     r"""Prediction input format for Image Object Detection.
     Attributes:
         content (str):
-            The image bytes or GCS URI to make the
-            prediction on.
+            The image bytes or Cloud Storage URI to make
+            the prediction on.
         mime_type (str):
             The MIME type of the content of the image.
             Only the images in below listed MIME types are

schema/predict/instance/google/cloud/aiplatform/v1beta1/schema/predict/instance_v1beta1/types/text_extraction.py

@@ -38,7 +38,7 @@ class TextExtractionPredictionInstance(proto.Message):
             If a key is provided, the batch prediction
             result will by mapped to this key. If omitted,
             then the batch prediction result will contain
-            the entire input instance. AI Platform will not
+            the entire input instance. Vertex AI will not
             check if keys in the request are duplicates, so
             it is up to the caller to ensure the keys are
             unique.

schema/predict/params/google/cloud/aiplatform/v1beta1/schema/predict/params_v1beta1/types/video_classification.py

@@ -39,16 +39,16 @@ class VideoClassificationPredictionParams(proto.Message):
             10,000.
         segment_classification (bool):
             Set to true to request segment-level
-            classification. AI Platform returns labels and
+            classification. Vertex AI returns labels and
             their confidence scores for the entire time
             segment of the video that user specified in the
             input instance. Default value is true
         shot_classification (bool):
             Set to true to request shot-level
-            classification. AI Platform determines the
+            classification. Vertex AI determines the
             boundaries for each camera shot in the entire
             time segment of the video that user specified in
-            the input instance. AI Platform then returns
+            the input instance. Vertex AI then returns
             labels and their confidence scores for each
             detected shot, along with the start and end time
             of the shot.
@@ -59,15 +59,14 @@ class VideoClassificationPredictionParams(proto.Message):
             Default value is false
         one_sec_interval_classification (bool):
             Set to true to request classification for a
-            video at one-second intervals. AI Platform
-            returns labels and their confidence scores for
-            each second of the entire time segment of the
-            video that user specified in the input WARNING:
-            Model evaluation is not done for this
-            classification type, the quality of it depends
-            on the training data, but there are no metrics
-            provided to describe that quality. Default value
-            is false
+            video at one-second intervals. Vertex AI returns
+            labels and their confidence scores for each
+            second of the entire time segment of the video
+            that user specified in the input WARNING: Model
+            evaluation is not done for this classification
+            type, the quality of it depends on the training
+            data, but there are no metrics provided to
+            describe that quality. Default value is false
     """
 
     confidence_threshold = proto.Field(

schema/predict/prediction/google/cloud/aiplatform/v1beta1/schema/predict/prediction_v1beta1/types/classification.py

@@ -29,8 +29,7 @@ class ClassificationPredictionResult(proto.Message):
     Attributes:
         ids (Sequence[int]):
             The resource IDs of the AnnotationSpecs that
-            had been identified, ordered by the confidence
-            score descendingly.
+            had been identified.
         display_names (Sequence[str]):
             The display names of the AnnotationSpecs that
             had been identified, order matches the IDs.

tests/unit/gapic/aiplatform_v1/test_migration_service.py

@@ -1700,18 +1700,20 @@ def test_parse_dataset_path():
 
 def test_dataset_path():
     project = "squid"
-    dataset = "clam"
-    expected = "projects/{project}/datasets/{dataset}".format(
-        project=project, dataset=dataset,
+    location = "clam"
+    dataset = "whelk"
+    expected = "projects/{project}/locations/{location}/datasets/{dataset}".format(
+        project=project, location=location, dataset=dataset,
     )
-    actual = MigrationServiceClient.dataset_path(project, dataset)
+    actual = MigrationServiceClient.dataset_path(project, location, dataset)
     assert expected == actual
 
 
 def test_parse_dataset_path():
     expected = {
-        "project": "whelk",
-        "dataset": "octopus",
+        "project": "octopus",
+        "location": "oyster",
+        "dataset": "nudibranch",
     }
     path = MigrationServiceClient.dataset_path(**expected)
 
@@ -1721,20 +1723,18 @@ def test_parse_dataset_path():
 
 
 def test_dataset_path():
-    project = "oyster"
-    location = "nudibranch"
-    dataset = "cuttlefish"
-    expected = "projects/{project}/locations/{location}/datasets/{dataset}".format(
-        project=project, location=location, dataset=dataset,
+    project = "cuttlefish"
+    dataset = "mussel"
+    expected = "projects/{project}/datasets/{dataset}".format(
+        project=project, dataset=dataset,
     )
-    actual = MigrationServiceClient.dataset_path(project, location, dataset)
+    actual = MigrationServiceClient.dataset_path(project, dataset)
     assert expected == actual
 
 
 def test_parse_dataset_path():
     expected = {
-        "project": "mussel",
-        "location": "winkle",
+        "project": "winkle",
         "dataset": "nautilus",
     }
     path = MigrationServiceClient.dataset_path(**expected)