feat: Added create_training_pipeline_custom_job_sample and create_tra…

…ining_pipeline_custom_training_managed_dataset_sample and fixed create_training_pipeline_image_classification_sample (#343)

* Added create_and_import_dataset_tabular_gcs_sample.py

* Added create_and_import_dataset_tabular_bigquery_sample.py

* Added create_training_pipeline_custom_job_sample.py and tweaked other tests

* Added create_training_pipeline_custom_training_managed_dataset_sample and fixed unmanaged sample

* Fixed args

* Deleted duplicated samples

* Added more args to samples

* Ran black

* Ran linter

Co-authored-by: sasha-gitg <44654632+sasha-gitg@users.noreply.github.com>

samples/model-builder/conftest.py

@@ -152,6 +152,19 @@ def mock_run_automl_image_training_job():
         yield mock
 
 
+@pytest.fixture
+def mock_init_custom_training_job():
+    with patch.object(aiplatform.training_jobs.CustomTrainingJob, "__init__") as mock:
+        mock.return_value = None
+        yield mock
+
+
+@pytest.fixture
+def mock_run_custom_training_job():
+    with patch.object(aiplatform.training_jobs.CustomTrainingJob, "run") as mock:
+        yield mock
+
+
 """
 ----------------------------------------------------------------------------
 Model Fixtures

samples/model-builder/create_training_pipeline_custom_job_sample.py

@@ -0,0 +1,69 @@
+# Copyright 2021 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from typing import List, Optional, Union
+
+from google.cloud import aiplatform
+
+
+#  [START aiplatform_sdk_create_training_pipeline_custom_job_sample]
+def create_training_pipeline_custom_job_sample(
+    project: str,
+    location: str,
+    display_name: str,
+    script_path: str,
+    container_uri: str,
+    model_serving_container_image_uri: str,
+    model_display_name: Optional[str] = None,
+    args: Optional[List[Union[str, float, int]]] = None,
+    replica_count: int = 0,
+    machine_type: str = "n1-standard-4",
+    accelerator_type: str = "ACCELERATOR_TYPE_UNSPECIFIED",
+    accelerator_count: int = 0,
+    training_fraction_split: float = 0.8,
+    validation_fraction_split: float = 0.1,
+    test_fraction_split: float = 0.1,
+    sync: bool = True,
+):
+    aiplatform.init(project=project, location=location)
+
+    job = aiplatform.CustomTrainingJob(
+        display_name=display_name,
+        script_path=script_path,
+        container_uri=container_uri,
+        model_serving_container_image_uri=model_serving_container_image_uri,
+    )
+
+    model = job.run(
+        model_display_name=model_display_name,
+        args=args,
+        replica_count=replica_count,
+        machine_type=machine_type,
+        accelerator_type=accelerator_type,
+        accelerator_count=accelerator_count,
+        training_fraction_split=training_fraction_split,
+        validation_fraction_split=validation_fraction_split,
+        test_fraction_split=test_fraction_split,
+        sync=sync,
+    )
+
+    model.wait()
+
+    print(model.display_name)
+    print(model.resource_name)
+    print(model.uri)
+    return model
+
+
+#  [END aiplatform_sdk_create_training_pipeline_custom_job_sample]

samples/model-builder/create_training_pipeline_custom_job_test.py

@@ -0,0 +1,62 @@
+# Copyright 2021 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+import create_training_pipeline_custom_job_sample
+import test_constants as constants
+
+
+def test_create_training_pipeline_custom_job_sample(
+    mock_sdk_init, mock_init_custom_training_job, mock_run_custom_training_job,
+):
+
+    create_training_pipeline_custom_job_sample.create_training_pipeline_custom_job_sample(
+        project=constants.PROJECT,
+        location=constants.LOCATION,
+        display_name=constants.DISPLAY_NAME,
+        args=constants.ARGS,
+        script_path=constants.SCRIPT_PATH,
+        container_uri=constants.CONTAINER_URI,
+        model_serving_container_image_uri=constants.CONTAINER_URI,
+        model_display_name=constants.DISPLAY_NAME_2,
+        replica_count=constants.REPLICA_COUNT,
+        machine_type=constants.MACHINE_TYPE,
+        accelerator_type=constants.ACCELERATOR_TYPE,
+        accelerator_count=constants.ACCELERATOR_COUNT,
+        training_fraction_split=constants.TRAINING_FRACTION_SPLIT,
+        validation_fraction_split=constants.VALIDATION_FRACTION_SPLIT,
+        test_fraction_split=constants.TEST_FRACTION_SPLIT,
+    )
+
+    mock_sdk_init.assert_called_once_with(
+        project=constants.PROJECT, location=constants.LOCATION
+    )
+    mock_init_custom_training_job.assert_called_once_with(
+        display_name=constants.DISPLAY_NAME,
+        script_path=constants.SCRIPT_PATH,
+        container_uri=constants.CONTAINER_URI,
+        model_serving_container_image_uri=constants.CONTAINER_URI,
+    )
+    mock_run_custom_training_job.assert_called_once_with(
+        model_display_name=constants.DISPLAY_NAME_2,
+        replica_count=constants.REPLICA_COUNT,
+        machine_type=constants.MACHINE_TYPE,
+        accelerator_type=constants.ACCELERATOR_TYPE,
+        accelerator_count=constants.ACCELERATOR_COUNT,
+        args=constants.ARGS,
+        training_fraction_split=constants.TRAINING_FRACTION_SPLIT,
+        validation_fraction_split=constants.VALIDATION_FRACTION_SPLIT,
+        test_fraction_split=constants.TEST_FRACTION_SPLIT,
+        sync=True,
+    )

samples/model-builder/create_training_pipeline_custom_training_managed_dataset_sample.py

@@ -0,0 +1,73 @@
+# Copyright 2021 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from typing import List, Optional, Union
+
+from google.cloud import aiplatform
+
+
+#  [START aiplatform_sdk_create_training_pipeline_custom_job_sample]
+def create_training_pipeline_custom_training_managed_dataset_sample(
+    project: str,
+    location: str,
+    display_name: str,
+    script_path: str,
+    container_uri: str,
+    model_serving_container_image_uri: str,
+    dataset_id: int,
+    model_display_name: Optional[str] = None,
+    args: Optional[List[Union[str, float, int]]] = None,
+    replica_count: int = 0,
+    machine_type: str = "n1-standard-4",
+    accelerator_type: str = "ACCELERATOR_TYPE_UNSPECIFIED",
+    accelerator_count: int = 0,
+    training_fraction_split: float = 0.8,
+    validation_fraction_split: float = 0.1,
+    test_fraction_split: float = 0.1,
+    sync: bool = True,
+):
+    aiplatform.init(project=project, location=location)
+
+    job = aiplatform.CustomTrainingJob(
+        display_name=display_name,
+        script_path=script_path,
+        container_uri=container_uri,
+        model_serving_container_image_uri=model_serving_container_image_uri,
+    )
+
+    my_image_ds = aiplatform.ImageDataset(dataset_id)
+
+    model = job.run(
+        dataset=my_image_ds,
+        model_display_name=model_display_name,
+        args=args,
+        replica_count=replica_count,
+        machine_type=machine_type,
+        accelerator_type=accelerator_type,
+        accelerator_count=accelerator_count,
+        training_fraction_split=training_fraction_split,
+        validation_fraction_split=validation_fraction_split,
+        test_fraction_split=test_fraction_split,
+        sync=sync,
+    )
+
+    model.wait()
+
+    print(model.display_name)
+    print(model.resource_name)
+    print(model.uri)
+    return model
+
+
+#  [END aiplatform_sdk_create_training_pipeline_custom_job_sample]

samples/model-builder/create_training_pipeline_custom_training_managed_dataset_test.py

@@ -0,0 +1,70 @@
+# Copyright 2021 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+import create_training_pipeline_custom_training_managed_dataset_sample
+import test_constants as constants
+
+
+def test_create_training_pipeline_custom_job_sample(
+    mock_sdk_init,
+    mock_image_dataset,
+    mock_init_custom_training_job,
+    mock_run_custom_training_job,
+    mock_get_image_dataset,
+):
+
+    create_training_pipeline_custom_training_managed_dataset_sample.create_training_pipeline_custom_training_managed_dataset_sample(
+        project=constants.PROJECT,
+        location=constants.LOCATION,
+        display_name=constants.DISPLAY_NAME,
+        args=constants.ARGS,
+        script_path=constants.SCRIPT_PATH,
+        container_uri=constants.CONTAINER_URI,
+        model_serving_container_image_uri=constants.CONTAINER_URI,
+        dataset_id=constants.RESOURCE_ID,
+        model_display_name=constants.DISPLAY_NAME_2,
+        replica_count=constants.REPLICA_COUNT,
+        machine_type=constants.MACHINE_TYPE,
+        accelerator_type=constants.ACCELERATOR_TYPE,
+        accelerator_count=constants.ACCELERATOR_COUNT,
+        training_fraction_split=constants.TRAINING_FRACTION_SPLIT,
+        validation_fraction_split=constants.VALIDATION_FRACTION_SPLIT,
+        test_fraction_split=constants.TEST_FRACTION_SPLIT,
+    )
+
+    mock_get_image_dataset.assert_called_once_with(constants.RESOURCE_ID)
+
+    mock_sdk_init.assert_called_once_with(
+        project=constants.PROJECT, location=constants.LOCATION
+    )
+    mock_init_custom_training_job.assert_called_once_with(
+        display_name=constants.DISPLAY_NAME,
+        script_path=constants.SCRIPT_PATH,
+        container_uri=constants.CONTAINER_URI,
+        model_serving_container_image_uri=constants.CONTAINER_URI,
+    )
+    mock_run_custom_training_job.assert_called_once_with(
+        dataset=mock_image_dataset,
+        model_display_name=constants.DISPLAY_NAME_2,
+        args=constants.ARGS,
+        replica_count=constants.REPLICA_COUNT,
+        machine_type=constants.MACHINE_TYPE,
+        accelerator_type=constants.ACCELERATOR_TYPE,
+        accelerator_count=constants.ACCELERATOR_COUNT,
+        training_fraction_split=constants.TRAINING_FRACTION_SPLIT,
+        validation_fraction_split=constants.VALIDATION_FRACTION_SPLIT,
+        test_fraction_split=constants.TEST_FRACTION_SPLIT,
+        sync=True,
+    )

samples/model-builder/create_training_pipeline_image_classification_sample.py

@@ -12,16 +12,18 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from typing import Optional
+
 from google.cloud import aiplatform
 
 
 #  [START aiplatform_sdk_create_training_pipeline_image_classification_sample]
 def create_training_pipeline_image_classification_sample(
     project: str,
+    location: str,
     display_name: str,
     dataset_id: int,
-    location: str = "us-central1",
-    model_display_name: str = None,
+    model_display_name: Optional[str] = None,
     training_fraction_split: float = 0.8,
     validation_fraction_split: float = 0.1,
     test_fraction_split: float = 0.1,

samples/model-builder/create_training_pipeline_image_classification_sample_test.py

@@ -27,6 +27,7 @@ def test_create_training_pipeline_image_classification_sample(
 
     create_training_pipeline_image_classification_sample.create_training_pipeline_image_classification_sample(
         project=constants.PROJECT,
+        location=constants.LOCATION,
         display_name=constants.DISPLAY_NAME,
         dataset_id=constants.RESOURCE_ID,
         model_display_name=constants.DISPLAY_NAME_2,

samples/model-builder/test_constants.py

@@ -40,6 +40,7 @@
 TRAINING_JOB_NAME = f"{PARENT}/trainingJobs/{RESOURCE_ID}"
 
 GCS_SOURCES = ["gs://bucket1/source1.jsonl", "gs://bucket7/source4.jsonl"]
+BIGQUERY_SOURCE = "bq://bigquery-public-data.ml_datasets.iris"
 GCS_DESTINATION = "gs://bucket3/output-dir/"
 
 TRAINING_FRACTION_SPLIT = 0.7
@@ -51,3 +52,11 @@
 ENCRYPTION_SPEC_KEY_NAME = f"{PARENT}/keyRings/{RESOURCE_ID}/cryptoKeys/{RESOURCE_ID_2}"
 
 PREDICTION_TEXT_INSTANCE = "This is some text for testing NLP prediction output"
+
+SCRIPT_PATH = "task.py"
+CONTAINER_URI = "gcr.io/my_project/my_image:latest"
+ARGS = ["--tfds", "tf_flowers:3.*.*"]
+REPLICA_COUNT = 0
+MACHINE_TYPE = "n1-standard-4"
+ACCELERATOR_TYPE = "ACCELERATOR_TYPE_UNSPECIFIED"
+ACCELERATOR_COUNT = 0