def setUp(self):
self.sagemaker = SageMakerTransformOperator(
task_id='test_sagemaker_operator',
aws_conn_id='sagemaker_test_id',
config=config,
wait_for_completion=False,
check_interval=5)
def test_pytorch_airflow_config_uploads_data_source_to_s3_when_inputs_not_provided(
sagemaker_session, cpu_instance_type):
with timeout(seconds=AIRFLOW_CONFIG_TIMEOUT_IN_SECONDS):
estimator = PyTorch(
entry_point=PYTORCH_MNIST_SCRIPT,
role=ROLE,
framework_version="1.1.0",
train_instance_count=2,
train_instance_type=cpu_instance_type,
hyperparameters={
"epochs": 6,
"backend": "gloo"
},
)
train_config = sm_airflow.training_config(estimator=estimator)
uploaded_s3_data = train_config["HyperParameters"][
"sagemaker_submit_directory"].strip('"')
transform_config = sm_airflow.transform_config_from_estimator(
estimator=estimator,
task_id="transform_config",
task_type="training",
instance_count=SINGLE_INSTANCE_COUNT,
instance_type=cpu_instance_type,
data=uploaded_s3_data,
content_type="text/csv",
)
default_args = {
"owner": "airflow",
"start_date": airflow.utils.dates.days_ago(2),
"provide_context": True,
}
dag = DAG("tensorflow_example",
default_args=default_args,
schedule_interval="@once")
train_op = SageMakerTrainingOperator(task_id="tf_training",
config=train_config,
wait_for_completion=True,
dag=dag)
transform_op = SageMakerTransformOperator(task_id="transform_operator",
config=transform_config,
wait_for_completion=True,
dag=dag)
transform_op.set_upstream(train_op)
_assert_that_s3_url_contains_data(sagemaker_session, uploaded_s3_data)
class TestSageMakerTransformOperator(unittest.TestCase):
def setUp(self):
configuration.load_test_config()
self.sagemaker = SageMakerTransformOperator(
task_id='test_sagemaker_operator',
aws_conn_id='sagemaker_test_id',
config=config,
wait_for_completion=False,
check_interval=5)
def test_parse_config_integers(self):
self.sagemaker.parse_config_integers()
test_config = self.sagemaker.config['Transform']
self.assertEqual(
test_config['TransformResources']['InstanceCount'],
int(test_config['TransformResources']['InstanceCount']))
self.assertEqual(test_config['MaxConcurrentTransforms'],
int(test_config['MaxConcurrentTransforms']))
self.assertEqual(test_config['MaxPayloadInMB'],
int(test_config['MaxPayloadInMB']))
@mock.patch.object(SageMakerHook, 'get_conn')
@mock.patch.object(SageMakerHook, 'create_model')
@mock.patch.object(SageMakerHook, 'create_transform_job')
def test_execute(self, mock_transform, mock_model, mock_client):
mock_transform.return_value = {
'TransformJobArn': 'testarn',
'ResponseMetadata': {
'HTTPStatusCode': 200
}
}
self.sagemaker.execute(None)
mock_model.assert_called_once_with(create_model_params)
mock_transform.assert_called_once_with(create_transform_params,
wait_for_completion=False,
check_interval=5,
max_ingestion_time=None)
@mock.patch.object(SageMakerHook, 'get_conn')
@mock.patch.object(SageMakerHook, 'create_model')
@mock.patch.object(SageMakerHook, 'create_transform_job')
def test_execute_with_failure(self, mock_transform, mock_model,
mock_client):
mock_transform.return_value = {
'TransformJobArn': 'testarn',
'ResponseMetadata': {
'HTTPStatusCode': 404
}
}
self.assertRaises(AirflowException, self.sagemaker.execute, None)
def _build_airflow_workflow(estimator,
instance_type,
inputs=None,
mini_batch_size=None):
training_config = sm_airflow.training_config(
estimator=estimator, inputs=inputs, mini_batch_size=mini_batch_size)
model = estimator.create_model()
assert model is not None
model_config = sm_airflow.model_config(instance_type, model)
assert model_config is not None
transform_config = sm_airflow.transform_config_from_estimator(
estimator=estimator,
task_id="transform_config",
task_type="training",
instance_count=SINGLE_INSTANCE_COUNT,
instance_type=estimator.train_instance_type,
data=inputs,
content_type="text/csv",
input_filter="$",
output_filter="$",
)
default_args = {
"owner": "airflow",
"start_date": airflow.utils.dates.days_ago(2),
"provide_context": True,
}
dag = DAG("tensorflow_example",
default_args=default_args,
schedule_interval="@once")
train_op = SageMakerTrainingOperator(task_id="tf_training",
config=training_config,
wait_for_completion=True,
dag=dag)
transform_op = SageMakerTransformOperator(task_id="transform_operator",
config=transform_config,
wait_for_completion=True,
dag=dag)
transform_op.set_upstream(train_op)
return training_config
class TestSageMakerTransformOperator(unittest.TestCase):
def setUp(self):
configuration.load_test_config()
self.sagemaker = SageMakerTransformOperator(
task_id='test_sagemaker_operator',
aws_conn_id='sagemaker_test_id',
config=config,
wait_for_completion=False,
check_interval=5
)
def test_parse_config_integers(self):
self.sagemaker.parse_config_integers()
test_config = self.sagemaker.config['Transform']
self.assertEqual(test_config['TransformResources']['InstanceCount'],
int(test_config['TransformResources']['InstanceCount']))
self.assertEqual(test_config['MaxConcurrentTransforms'],
int(test_config['MaxConcurrentTransforms']))
self.assertEqual(test_config['MaxPayloadInMB'],
int(test_config['MaxPayloadInMB']))
@mock.patch.object(SageMakerHook, 'get_conn')
@mock.patch.object(SageMakerHook, 'create_model')
@mock.patch.object(SageMakerHook, 'create_transform_job')
def test_execute(self, mock_transform, mock_model, mock_client):
mock_transform.return_value = {'TransformJobArn': 'testarn',
'ResponseMetadata':
{'HTTPStatusCode': 200}}
self.sagemaker.execute(None)
mock_model.assert_called_once_with(create_model_params)
mock_transform.assert_called_once_with(create_transform_params,
wait_for_completion=False,
check_interval=5,
max_ingestion_time=None
)
@mock.patch.object(SageMakerHook, 'get_conn')
@mock.patch.object(SageMakerHook, 'create_model')
@mock.patch.object(SageMakerHook, 'create_transform_job')
def test_execute_with_failure(self, mock_transform, mock_model, mock_client):
mock_transform.return_value = {'TransformJobArn': 'testarn',
'ResponseMetadata':
{'HTTPStatusCode': 404}}
self.assertRaises(AirflowException, self.sagemaker.execute, None)
def setUp(self):
configuration.load_test_config()
self.sagemaker = SageMakerTransformOperator(
task_id='test_sagemaker_operator',
aws_conn_id='sagemaker_test_id',
config=config,
wait_for_completion=False,
check_interval=5
)
wait_for_completion=True,
check_interval=30)
# launch sagemaker hyperparameter job and wait until it completes
tune_model_task = SageMakerTuningOperator(task_id='model_tuning',
dag=dag,
config=tuner_config,
aws_conn_id='airflow-sagemaker',
wait_for_completion=True,
check_interval=30)
# launch sagemaker batch transform job and wait until it completes
batch_transform_task = SageMakerTransformOperator(
task_id='predicting',
dag=dag,
config=transform_config,
aws_conn_id='airflow-sagemaker',
wait_for_completion=True,
check_interval=30,
trigger_rule=TriggerRule.ONE_SUCCESS)
basher_task = BashOperator(task_id='sleep', bash_command='sleep 5', dag=dag)
cleanup_task = DummyOperator(task_id='cleaning_up', dag=dag)
# set the dependencies between tasks
init >> preprocess_task
preprocess_task >> prepare_task
prepare_task >> branching
branching >> tune_model_task
branching >> train_model_task
'role': role,
'sess': sess,
'spark_model_uri':
config['inference_pipeline']['inputs']['spark_model'],
'pipeline_model_name':
config['inference_pipeline']['pipeline_model_name'],
'endpoint_name': config['inference_pipeline']['pipeline_model_name'],
'region': region,
'bucket': bucket
})
# launch sagemaker batch transform job and wait until it completes
batch_transform_task = SageMakerTransformOperator(
task_id='batch_predicting',
dag=dag,
config=transform_config,
aws_conn_id='airflow-sagemaker',
wait_for_completion=True,
check_interval=30)
# Cleanup task, deletes ALL SageMaker endpoints and model artifacts
# Uncomment below clean_up_task to clean up sagemaker endpoint resources and model artifacts
# clean_up_task = PythonOperator(
# task_id='clean_up',
# dag=dag,
# python_callable=clean_up.clean_up,
# op_kwargs={'region': region, "bucket": bucket}
# )
init.set_downstream(sm_proc_job_task)
# launch sagemaker hyperparameter job and wait until it completes
tune_model_task = SageMakerTuningOperator(
task_id="model_tuning",
dag=dag,
config=tuner_config,
aws_conn_id="airflow-sagemaker",
wait_for_completion=True,
check_interval=30,
)
# launch sagemaker batch transform job and wait until it completes
batch_transform_task = SageMakerTransformOperator(
task_id="predicting",
dag=dag,
config=transform_config,
aws_conn_id="airflow-sagemaker",
wait_for_completion=True,
check_interval=30,
trigger_rule=TriggerRule.ONE_SUCCESS,
)
cleanup_task = DummyOperator(task_id="cleaning_up", dag=dag)
# set the dependencies between tasks
init.set_downstream(preprocess_task)
preprocess_task.set_downstream(prepare_task)
prepare_task.set_downstream(branching)
branching.set_downstream(tune_model_task)
branching.set_downstream(train_model_task)
tune_model_task.set_downstream(batch_transform_task)
