feat: add video action recognition samples (#77)

Co-authored-by: Yu-Han Liu <yuhanliu@google.com>

Author: morgandu

samples/snippets/create_batch_prediction_job_video_action_recognition_sample.py

@@ -0,0 +1,60 @@
+# Copyright 2020 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.
+
+# [START aiplatform_create_batch_prediction_job_video_action_recognition_sample]
+from google.cloud import aiplatform
+from google.protobuf import json_format
+from google.protobuf.struct_pb2 import Value
+
+
+def create_batch_prediction_job_video_action_recognition_sample(
+    project: str,
+    display_name: str,
+    model: str,
+    gcs_source_uri: str,
+    gcs_destination_output_uri_prefix: str,
+    location: str = "us-central1",
+    api_endpoint: str = "us-central1-aiplatform.googleapis.com",
+):
+    client_options = {"api_endpoint": api_endpoint}
+    # Initialize client that will be used to create and send requests.
+    # This client only needs to be created once, and can be reused for multiple requests.
+    client = aiplatform.gapic.JobServiceClient(client_options=client_options)
+    model_parameters_dict = {
+        "confidenceThreshold": 0.5,
+    }
+    model_parameters = json_format.ParseDict(model_parameters_dict, Value())
+
+    batch_prediction_job = {
+        "display_name": display_name,
+        # Format: 'projects/{project}/locations/{location}/models/{model_id}'
+        "model": model,
+        "model_parameters": model_parameters,
+        "input_config": {
+            "instances_format": "jsonl",
+            "gcs_source": {"uris": [gcs_source_uri]},
+        },
+        "output_config": {
+            "predictions_format": "jsonl",
+            "gcs_destination": {"output_uri_prefix": gcs_destination_output_uri_prefix},
+        },
+    }
+    parent = f"projects/{project}/locations/{location}"
+    response = client.create_batch_prediction_job(
+        parent=parent, batch_prediction_job=batch_prediction_job
+    )
+    print("response:", response)
+
+
+# [END aiplatform_create_batch_prediction_job_video_action_recognition_sample]

samples/snippets/create_batch_prediction_job_video_action_recognition_test.py

@@ -0,0 +1,82 @@
+# Copyright 2020 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 uuid
+import pytest
+import os
+
+import helpers
+
+import create_batch_prediction_job_video_action_recognition_sample
+
+from google.cloud import aiplatform
+
+PROJECT_ID = os.getenv("BUILD_SPECIFIC_GCLOUD_PROJECT")
+LOCATION = "us-central1"
+MODEL_ID = "3530998029718913024"  # permanent_swim_run_videos_action_recognition_model
+DISPLAY_NAME = f"temp_create_batch_prediction_job_video_action_recognition_test_{uuid.uuid4()}"
+GCS_SOURCE_URI = "gs://automl-video-demo-data/ucaip-var/swimrun_bp.jsonl"
+GCS_OUTPUT_URI = "gs://ucaip-samples-test-output/"
+API_ENDPOINT = "us-central1-aiplatform.googleapis.com"
+
+@pytest.fixture
+def shared_state():
+    state = {}
+    yield state
+
+
+@pytest.fixture
+def job_client():
+    client_options = {"api_endpoint": API_ENDPOINT}
+    job_client = aiplatform.gapic.JobServiceClient(
+        client_options=client_options)
+    yield job_client
+
+
+@pytest.fixture(scope="function", autouse=True)
+def teardown(shared_state, job_client):
+    yield
+    job_client.delete_batch_prediction_job(
+        name=shared_state["batch_prediction_job_name"]
+    )
+
+
+# Creating AutoML Video Object Tracking batch prediction job
+def test_create_batch_prediction_job_video_action_recognition_sample(
+    capsys, shared_state, job_client
+):
+
+    model = f"projects/{PROJECT_ID}/locations/{LOCATION}/models/{MODEL_ID}"
+
+    create_batch_prediction_job_video_action_recognition_sample.create_batch_prediction_job_video_action_recognition_sample(
+        project=PROJECT_ID,
+        display_name=DISPLAY_NAME,
+        model=model,
+        gcs_source_uri=GCS_SOURCE_URI,
+        gcs_destination_output_uri_prefix=GCS_OUTPUT_URI,
+    )
+
+    out, _ = capsys.readouterr()
+
+    # Save resource name of the newly created batch prediction job
+    shared_state["batch_prediction_job_name"] = helpers.get_name(out)
+
+    # Waiting for batch prediction job to be in CANCELLED state
+    helpers.wait_for_job_state(
+        get_job_method=job_client.get_batch_prediction_job,
+        name=shared_state["batch_prediction_job_name"],
+        expected_state="SUCCEEDED",
+        timeout=600,
+        freq=20,
+    )

samples/snippets/create_training_pipeline_video_action_recognition_sample.py

@@ -0,0 +1,54 @@
+# Copyright 2020 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.
+
+# [START aiplatform_create_training_pipeline_video_action_recognition_sample]
+from google.cloud import aiplatform
+from google.protobuf import json_format
+from google.protobuf.struct_pb2 import Value
+
+
+def create_training_pipeline_video_action_recognition_sample(
+    project: str,
+    display_name: str,
+    dataset_id: str,
+    model_display_name: str,
+    model_type: str,
+    location: str = "us-central1",
+    api_endpoint: str = "us-central1-aiplatform.googleapis.com",
+):
+    client_options = {"api_endpoint": api_endpoint}
+    # Initialize client that will be used to create and send requests.
+    # This client only needs to be created once, and can be reused for multiple requests.
+    client = aiplatform.gapic.PipelineServiceClient(client_options=client_options)
+    training_task_inputs_dict = {
+        # modelType can be either 'CLOUD' or 'MOBILE_VERSATILE_1'
+        "modelType": model_type
+    }
+    training_task_inputs = json_format.ParseDict(training_task_inputs_dict, Value())
+
+    training_pipeline = {
+        "display_name": display_name,
+        "training_task_definition": "gs://google-cloud-aiplatform/schema/trainingjob/definition/automl_video_action_recognition_1.0.0.yaml",
+        "training_task_inputs": training_task_inputs,
+        "input_data_config": {"dataset_id": dataset_id},
+        "model_to_upload": {"display_name": model_display_name},
+    }
+    parent = f"projects/{project}/locations/{location}"
+    response = client.create_training_pipeline(
+        parent=parent, training_pipeline=training_pipeline
+    )
+    print("response:", response)
+
+
+# [END aiplatform_create_training_pipeline_video_action_recognition_sample]

samples/snippets/create_training_pipeline_video_action_recognition_test.py

@@ -0,0 +1,100 @@
+# Copyright 2020 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 uuid
+import pytest
+import os
+
+import helpers
+
+import create_training_pipeline_video_action_recognition_sample
+
+from google.cloud import aiplatform
+
+LOCATION = "us-central1"
+PROJECT_ID = os.getenv("BUILD_SPECIFIC_GCLOUD_PROJECT")
+DATASET_ID = "6881957627459272704"  # permanent_swim_run_videos_action_recognition_dataset
+DISPLAY_NAME = f"temp_create_training_pipeline_video_action_recognition_test_{uuid.uuid4()}"
+MODEL_DISPLAY_NAME = f"Temp Model for {DISPLAY_NAME}"
+MODEL_TYPE = "CLOUD"
+API_ENDPOINT = "us-central1-aiplatform.googleapis.com"
+
+@pytest.fixture
+def shared_state():
+    state = {}
+    yield state
+
+
+@pytest.fixture
+def pipeline_client():
+    client_options = {"api_endpoint": API_ENDPOINT}
+    pipeline_client = aiplatform.gapic.PipelineServiceClient(
+        client_options=client_options
+    )
+    yield pipeline_client
+
+
+@pytest.fixture
+def model_client():
+    client_options = {"api_endpoint": API_ENDPOINT}
+    model_client = aiplatform.gapic.ModelServiceClient(
+        client_options=client_options)
+    yield model_client
+
+
+@pytest.fixture(scope="function", autouse=True)
+def teardown(shared_state, model_client, pipeline_client):
+    yield
+    model_client.delete_model(name=shared_state["model_name"])
+    pipeline_client.delete_training_pipeline(
+        name=shared_state["training_pipeline_name"]
+    )
+
+
+# Training AutoML Vision Model
+def test_create_training_pipeline_video_action_recognition_sample(
+    capsys, shared_state, pipeline_client
+):
+    create_training_pipeline_video_action_recognition_sample.create_training_pipeline_video_action_recognition_sample(
+        project=PROJECT_ID,
+        display_name=DISPLAY_NAME,
+        dataset_id=DATASET_ID,
+        model_display_name=MODEL_DISPLAY_NAME,
+        model_type=MODEL_TYPE,
+    )
+
+    out, _ = capsys.readouterr()
+
+    assert "response:" in out
+
+    # Save resource name of the newly created training pipeline
+    shared_state["training_pipeline_name"] = helpers.get_name(out)
+
+    # Poll until the pipeline succeeds because we want to test the model_upload step as well.
+    helpers.wait_for_job_state(
+        get_job_method=pipeline_client.get_training_pipeline,
+        name=shared_state["training_pipeline_name"],
+        expected_state="SUCCEEDED",
+        timeout=5000,
+        freq=20,
+    )
+
+    training_pipeline = pipeline_client.get_training_pipeline(
+        name=shared_state["training_pipeline_name"]
+    )
+
+    # Check that the model indeed has been uploaded.
+    assert training_pipeline.model_to_upload.name != ""
+
+    shared_state["model_name"] = training_pipeline.model_to_upload.name

samples/snippets/export_model_video_action_recognition_sample.py

@@ -0,0 +1,44 @@
+# Copyright 2020 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.
+
+# [START aiplatform_export_model_video_action_recognition_sample]
+from google.cloud import aiplatform
+
+
+def export_model_video_action_recognition_sample(
+    project: str,
+    model_id: str,
+    gcs_destination_output_uri_prefix: str,
+    export_format: str,
+    location: str = "us-central1",
+    api_endpoint: str = "us-central1-aiplatform.googleapis.com",
+    timeout: int = 300,
+):
+    client_options = {"api_endpoint": api_endpoint}
+    # Initialize client that will be used to create and send requests.
+    # This client only needs to be created once, and can be reused for multiple requests.
+    client = aiplatform.gapic.ModelServiceClient(client_options=client_options)
+    gcs_destination = {"output_uri_prefix": gcs_destination_output_uri_prefix}
+    output_config = {
+        "artifact_destination": gcs_destination,
+        "export_format_id": export_format,
+    }
+    name = client.model_path(project=project, location=location, model=model_id)
+    response = client.export_model(name=name, output_config=output_config)
+    print("Long running operation:", response.operation.name)
+    export_model_response = response.result(timeout=timeout)
+    print("export_model_response:", export_model_response)
+
+
+# [END aiplatform_export_model_video_action_recognition_sample]

samples/snippets/export_model_video_action_recognition_test.py

@@ -0,0 +1,45 @@
+# Copyright 2020 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 pytest
+import os
+
+import export_model_video_action_recognition_sample
+from google.cloud import storage
+
+PROJECT_ID = os.getenv("BUILD_SPECIFIC_GCLOUD_PROJECT")
+MODEL_ID = "3422489426196955136"  # permanent_swim_run_videos_action_recognition_edge_model
+GCS_URI = "gs://ucaip-samples-test-output/tmp/export_model_video_action_recognition_sample"
+EXPORT_FORMAT = "tf-saved-model"
+
+@pytest.fixture(scope="function", autouse=True)
+def teardown():
+    yield
+
+    storage_client = storage.Client()
+    bucket = storage_client.get_bucket("ucaip-samples-test-output")
+    blobs = bucket.list_blobs(prefix="tmp/export_model_video_action_recognition_sample")
+    for blob in blobs:
+        blob.delete()
+
+
+def test_export_model_video_action_recognition_sample(capsys):
+    export_model_video_action_recognition_sample.export_model_video_action_recognition_sample(
+        project=PROJECT_ID,
+        model_id=MODEL_ID,
+        gcs_destination_output_uri_prefix=GCS_URI,
+        export_format=EXPORT_FORMAT,
+    )
+    out, _ = capsys.readouterr()
+    assert "export_model_response" in out