def __init__(self,
name: str,
model: Model,
inputs: CreateModelInput,
depends_on: List[str] = None):
"""Construct a CreateModelStep, given an `sagemaker.model.Model` instance.
In addition to the Model instance, the other arguments are those that are supplied to
the `_create_sagemaker_model` method of the `sagemaker.model.Model._create_sagemaker_model`.
Args:
name (str): The name of the CreateModel step.
model (Model): A `sagemaker.model.Model` instance.
inputs (CreateModelInput): A `sagemaker.inputs.CreateModelInput` instance.
Defaults to `None`.
depends_on (List[str]): A list of step names this `sagemaker.workflow.steps.CreateModelStep`
depends on
"""
super(CreateModelStep, self).__init__(name, StepTypeEnum.CREATE_MODEL,
depends_on)
self.model = model
self.inputs = inputs or CreateModelInput()
self._properties = Properties(path=f"Steps.{name}",
shape_name="DescribeModelOutput")
def test_create_model_step_with_model_pipeline(tfo, time, sagemaker_session):
framework_model = DummyFrameworkModel(sagemaker_session)
sparkml_model = SparkMLModel(
model_data="s3://bucket/model_2.tar.gz",
role=ROLE,
sagemaker_session=sagemaker_session,
env={"SAGEMAKER_DEFAULT_INVOCATIONS_ACCEPT": "text/csv"},
)
model = PipelineModel(models=[framework_model, sparkml_model],
role=ROLE,
sagemaker_session=sagemaker_session)
inputs = CreateModelInput(
instance_type="c4.4xlarge",
accelerator_type="ml.eia1.medium",
)
step = CreateModelStep(
name="MyCreateModelStep",
depends_on=["TestStep"],
display_name="MyCreateModelStep",
description="TestDescription",
model=model,
inputs=inputs,
)
step.add_depends_on(["SecondTestStep"])
assert step.to_request() == {
"Name": "MyCreateModelStep",
"Type": "Model",
"Description": "TestDescription",
"DisplayName": "MyCreateModelStep",
"DependsOn": ["TestStep", "SecondTestStep"],
"Arguments": {
"Containers": [
{
"Environment": {
"SAGEMAKER_PROGRAM": "dummy_script.py",
"SAGEMAKER_SUBMIT_DIRECTORY":
"s3://my-bucket/mi-1-2017-10-10-14-14-15/sourcedir.tar.gz",
"SAGEMAKER_CONTAINER_LOG_LEVEL": "20",
"SAGEMAKER_REGION": "us-west-2",
},
"Image": "mi-1",
"ModelDataUrl": "s3://bucket/model_1.tar.gz",
},
{
"Environment": {
"SAGEMAKER_DEFAULT_INVOCATIONS_ACCEPT": "text/csv"
},
"Image":
"246618743249.dkr.ecr.us-west-2.amazonaws.com/sagemaker-sparkml-serving:2.4",
"ModelDataUrl": "s3://bucket/model_2.tar.gz",
},
],
"ExecutionRoleArn":
"DummyRole",
},
}
assert step.properties.ModelName.expr == {
"Get": "Steps.MyCreateModelStep.ModelName"
}
def test_create_model_step(sagemaker_session):
model = Model(
image_uri=IMAGE_URI,
role=ROLE,
sagemaker_session=sagemaker_session,
)
inputs = CreateModelInput(
instance_type="c4.4xlarge",
accelerator_type="ml.eia1.medium",
)
step = CreateModelStep(
name="MyCreateModelStep",
depends_on=["TestStep"],
model=model,
inputs=inputs,
)
step.add_depends_on(["SecondTestStep"])
assert step.to_request() == {
"Name": "MyCreateModelStep",
"Type": "Model",
"DependsOn": ["TestStep", "SecondTestStep"],
"Arguments": {
"ExecutionRoleArn": "DummyRole",
"PrimaryContainer": {
"Environment": {},
"Image": "fakeimage"
},
},
}
assert step.properties.ModelName.expr == {
"Get": "Steps.MyCreateModelStep.ModelName"
}
def __init__(
self,
name: str,
model: Union[Model, PipelineModel],
inputs: CreateModelInput = None,
depends_on: Union[List[str], List[Step]] = None,
retry_policies: List[RetryPolicy] = None,
display_name: str = None,
description: str = None,
):
"""Construct a `CreateModelStep`, given an `sagemaker.model.Model` instance.
In addition to the `Model` instance, the other arguments are those that are supplied to
the `_create_sagemaker_model` method of the `sagemaker.model.Model._create_sagemaker_model`.
Args:
name (str): The name of the `CreateModelStep`.
model (Model or PipelineModel): A `sagemaker.model.Model`
or `sagemaker.pipeline.PipelineModel` instance.
inputs (CreateModelInput): A `sagemaker.inputs.CreateModelInput` instance.
Defaults to `None`.
depends_on (List[str] or List[Step]): A list of `Step` names or `Step` instances
this `sagemaker.workflow.steps.CreateModelStep` depends on.
retry_policies (List[RetryPolicy]): A list of retry policies.
display_name (str): The display name of the `CreateModelStep`.
description (str): The description of the `CreateModelStep`.
"""
super(CreateModelStep,
self).__init__(name, StepTypeEnum.CREATE_MODEL, display_name,
description, depends_on, retry_policies)
self.model = model
self.inputs = inputs or CreateModelInput()
self._properties = Properties(path=f"Steps.{name}",
shape_name="DescribeModelOutput")
def test_estimator_transformer(estimator):
model_data = f"s3://{BUCKET}/model.tar.gz"
model_inputs = CreateModelInput(
instance_type="c4.4xlarge",
accelerator_type="ml.eia1.medium",
)
transform_inputs = TransformInput(data=f"s3://{BUCKET}/transform_manifest")
estimator_transformer = EstimatorTransformer(
name="EstimatorTransformerStep",
estimator=estimator,
model_data=model_data,
model_inputs=model_inputs,
instance_count=1,
instance_type="ml.c4.4xlarge",
transform_inputs=transform_inputs,
)
request_dicts = estimator_transformer.request_dicts()
assert len(request_dicts) == 2
for request_dict in request_dicts:
if request_dict["Type"] == "Model":
assert request_dict == {
"Name": "EstimatorTransformerStepCreateModelStep",
"Type": "Model",
"Arguments": {
"ExecutionRoleArn": "DummyRole",
"PrimaryContainer": {
"Environment": {},
"Image": "fakeimage",
"ModelDataUrl": "s3://my-bucket/model.tar.gz",
},
},
}
elif request_dict["Type"] == "Transform":
assert request_dict[
"Name"] == "EstimatorTransformerStepTransformStep"
arguments = request_dict["Arguments"]
assert isinstance(arguments["ModelName"], Properties)
arguments.pop("ModelName")
assert arguments == {
"TransformInput": {
"DataSource": {
"S3DataSource": {
"S3DataType": "S3Prefix",
"S3Uri": f"s3://{BUCKET}/transform_manifest",
}
}
},
"TransformOutput": {
"S3OutputPath": None
},
"TransformResources": {
"InstanceCount": 1,
"InstanceType": "ml.c4.4xlarge"
},
}
else:
raise Exception(
"A step exists in the collection of an invalid type.")
def __init__(
self,
name: str,
step_args: Optional[dict] = None,
model: Optional[Union[Model, PipelineModel]] = None,
inputs: Optional[CreateModelInput] = None,
depends_on: Optional[List[Union[str, Step, "StepCollection"]]] = None,
retry_policies: Optional[List[RetryPolicy]] = None,
display_name: Optional[str] = None,
description: Optional[str] = None,
):
"""Construct a `CreateModelStep`, given an `sagemaker.model.Model` instance.
In addition to the `Model` instance, the other arguments are those that are supplied to
the `_create_sagemaker_model` method of the `sagemaker.model.Model._create_sagemaker_model`.
Args:
name (str): The name of the `CreateModelStep`.
step_args (dict): The arguments for the `CreateModelStep` definition (default: None).
model (Model or PipelineModel): A `sagemaker.model.Model`
or `sagemaker.pipeline.PipelineModel` instance (default: None).
inputs (CreateModelInput): A `sagemaker.inputs.CreateModelInput` instance.
(default: None).
depends_on (List[Union[str, Step, StepCollection]]): A list of `Step`/`StepCollection`
names or `Step` instances or `StepCollection` instances that this `CreateModelStep`
depends on (default: None).
retry_policies (List[RetryPolicy]): A list of retry policies (default: None).
display_name (str): The display name of the `CreateModelStep` (default: None).
description (str): The description of the `CreateModelStep` (default: None).
"""
super(CreateModelStep,
self).__init__(name, StepTypeEnum.CREATE_MODEL, display_name,
description, depends_on, retry_policies)
if not (step_args is None) ^ (model is None):
raise ValueError(
"step_args and model are mutually exclusive. Either of them should be provided."
)
self.step_args = step_args
self.model = model
self.inputs = inputs or CreateModelInput()
self._properties = Properties(path=f"Steps.{name}",
shape_name="DescribeModelOutput")
# TODO: add public document link here once ready
warnings.warn(
("We are deprecating the use of CreateModelStep. "
"Instead, please use the ModelStep, which simply takes in the step arguments "
"generated by model.create()."),
DeprecationWarning,
)
def test_model_registration_with_model_repack(
sagemaker_session,
role,
pipeline_name,
region_name,
):
base_dir = os.path.join(DATA_DIR, "pytorch_mnist")
entry_point = os.path.join(base_dir, "mnist.py")
input_path = sagemaker_session.upload_data(
path=os.path.join(base_dir, "training"),
key_prefix="integ-test-data/pytorch_mnist/training",
)
inputs = TrainingInput(s3_data=input_path)
instance_count = ParameterInteger(name="InstanceCount", default_value=1)
instance_type = ParameterString(name="InstanceType",
default_value="ml.m5.xlarge")
good_enough_input = ParameterInteger(name="GoodEnoughInput",
default_value=1)
pytorch_estimator = PyTorch(
entry_point=entry_point,
role=role,
framework_version="1.5.0",
py_version="py3",
instance_count=instance_count,
instance_type=instance_type,
sagemaker_session=sagemaker_session,
)
step_train = TrainingStep(
name="pytorch-train",
estimator=pytorch_estimator,
inputs=inputs,
)
step_register = RegisterModel(
name="pytorch-register-model",
estimator=pytorch_estimator,
model_data=step_train.properties.ModelArtifacts.S3ModelArtifacts,
content_types=["*"],
response_types=["*"],
inference_instances=["*"],
transform_instances=["*"],
description="test-description",
entry_point=entry_point,
)
model = Model(
image_uri=pytorch_estimator.training_image_uri(),
model_data=step_train.properties.ModelArtifacts.S3ModelArtifacts,
sagemaker_session=sagemaker_session,
role=role,
)
model_inputs = CreateModelInput(
instance_type="ml.m5.large",
accelerator_type="ml.eia1.medium",
)
step_model = CreateModelStep(
name="pytorch-model",
model=model,
inputs=model_inputs,
)
step_cond = ConditionStep(
name="cond-good-enough",
conditions=[
ConditionGreaterThanOrEqualTo(left=good_enough_input, right=1)
],
if_steps=[step_train, step_register],
else_steps=[step_model],
)
pipeline = Pipeline(
name=pipeline_name,
parameters=[good_enough_input, instance_count, instance_type],
steps=[step_cond],
sagemaker_session=sagemaker_session,
)
try:
response = pipeline.create(role)
create_arn = response["PipelineArn"]
assert re.match(
fr"arn:aws:sagemaker:{region_name}:\d{{12}}:pipeline/{pipeline_name}",
create_arn)
execution = pipeline.start(parameters={})
assert re.match(
fr"arn:aws:sagemaker:{region_name}:\d{{12}}:pipeline/{pipeline_name}/execution/",
execution.arn,
)
execution = pipeline.start(parameters={"GoodEnoughInput": 0})
assert re.match(
fr"arn:aws:sagemaker:{region_name}:\d{{12}}:pipeline/{pipeline_name}/execution/",
execution.arn,
)
finally:
try:
pipeline.delete()
except Exception:
pass
def test_three_step_definition(
sagemaker_session,
region_name,
role,
script_dir,
pipeline_name,
athena_dataset_definition,
):
framework_version = "0.20.0"
instance_type = ParameterString(name="InstanceType",
default_value="ml.m5.xlarge")
instance_count = ParameterInteger(name="InstanceCount", default_value=1)
output_prefix = ParameterString(name="OutputPrefix",
default_value="output")
input_data = f"s3://sagemaker-sample-data-{region_name}/processing/census/census-income.csv"
sklearn_processor = SKLearnProcessor(
framework_version=framework_version,
instance_type=instance_type,
instance_count=instance_count,
base_job_name="test-sklearn",
sagemaker_session=sagemaker_session,
role=role,
)
step_process = ProcessingStep(
name="my-process",
processor=sklearn_processor,
inputs=[
ProcessingInput(source=input_data,
destination="/opt/ml/processing/input"),
ProcessingInput(dataset_definition=athena_dataset_definition),
],
outputs=[
ProcessingOutput(output_name="train_data",
source="/opt/ml/processing/train"),
ProcessingOutput(
output_name="test_data",
source="/opt/ml/processing/test",
destination=Join(
on="/",
values=[
"s3:/",
sagemaker_session.default_bucket(),
"test-sklearn",
output_prefix,
ExecutionVariables.PIPELINE_EXECUTION_ID,
],
),
),
],
code=os.path.join(script_dir, "preprocessing.py"),
)
sklearn_train = SKLearn(
framework_version=framework_version,
entry_point=os.path.join(script_dir, "train.py"),
instance_type=instance_type,
sagemaker_session=sagemaker_session,
role=role,
)
step_train = TrainingStep(
name="my-train",
estimator=sklearn_train,
inputs=TrainingInput(
s3_data=step_process.properties.ProcessingOutputConfig.
Outputs["train_data"].S3Output.S3Uri),
)
model = Model(
image_uri=sklearn_train.image_uri,
model_data=step_train.properties.ModelArtifacts.S3ModelArtifacts,
sagemaker_session=sagemaker_session,
role=role,
)
model_inputs = CreateModelInput(
instance_type="ml.m5.large",
accelerator_type="ml.eia1.medium",
)
step_model = CreateModelStep(
name="my-model",
model=model,
inputs=model_inputs,
)
pipeline = Pipeline(
name=pipeline_name,
parameters=[instance_type, instance_count, output_prefix],
steps=[step_process, step_train, step_model],
sagemaker_session=sagemaker_session,
)
definition = json.loads(pipeline.definition())
assert definition["Version"] == "2020-12-01"
assert set(tuple(param.items())
for param in definition["Parameters"]) == set([
tuple({
"Name": "InstanceType",
"Type": "String",
"DefaultValue": "ml.m5.xlarge"
}.items()),
tuple({
"Name": "InstanceCount",
"Type": "Integer",
"DefaultValue": 1
}.items()),
tuple({
"Name": "OutputPrefix",
"Type": "String",
"DefaultValue": "output"
}.items()),
])
steps = definition["Steps"]
assert len(steps) == 3
names_and_types = []
processing_args = {}
training_args = {}
for step in steps:
names_and_types.append((step["Name"], step["Type"]))
if step["Type"] == "Processing":
processing_args = step["Arguments"]
if step["Type"] == "Training":
training_args = step["Arguments"]
if step["Type"] == "Model":
model_args = step["Arguments"]
assert set(names_and_types) == set([
("my-process", "Processing"),
("my-train", "Training"),
("my-model", "Model"),
])
assert processing_args["ProcessingResources"]["ClusterConfig"] == {
"InstanceType": {
"Get": "Parameters.InstanceType"
},
"InstanceCount": {
"Get": "Parameters.InstanceCount"
},
"VolumeSizeInGB": 30,
}
assert training_args["ResourceConfig"] == {
"InstanceCount": 1,
"InstanceType": {
"Get": "Parameters.InstanceType"
},
"VolumeSizeInGB": 30,
}
assert training_args["InputDataConfig"][0]["DataSource"]["S3DataSource"][
"S3Uri"] == {
"Get":
"Steps.my-process.ProcessingOutputConfig.Outputs['train_data'].S3Output.S3Uri"
}
assert model_args["PrimaryContainer"]["ModelDataUrl"] == {
"Get": "Steps.my-train.ModelArtifacts.S3ModelArtifacts"
}
try:
response = pipeline.create(role)
create_arn = response["PipelineArn"]
assert re.match(
fr"arn:aws:sagemaker:{region_name}:\d{{12}}:pipeline/{pipeline_name}",
create_arn,
)
finally:
try:
pipeline.delete()
except Exception:
pass
def get_pipeline(region, role, default_bucket, pipeline_name,
model_package_group_name, base_job_prefix):
"""Gets a SageMaker ML Pipeline instance working with BERT.
Args:
region: AWS region to create and run the pipeline.
role: IAM role to create and run steps and pipeline.
default_bucket: the bucket to use for storing the artifacts
pipeline_name: name of this pipeline
model_package_group_name: model package group
base_job_prefix: prefic of the job name
Returns:
an instance of a pipeline
"""
sm = boto3.Session().client(service_name="sagemaker", region_name=region)
input_data = ParameterString(
name="InputDataUrl",
default_value="s3://{}/amazon-reviews-pds/tsv/".format(bucket),
)
processing_instance_count = ParameterInteger(
name="ProcessingInstanceCount", default_value=1)
processing_instance_type = ParameterString(name="ProcessingInstanceType",
default_value="ml.c5.2xlarge")
max_seq_length = ParameterInteger(
name="MaxSeqLength",
default_value=64,
)
balance_dataset = ParameterString(
name="BalanceDataset",
default_value="True",
)
train_split_percentage = ParameterFloat(
name="TrainSplitPercentage",
default_value=0.90,
)
validation_split_percentage = ParameterFloat(
name="ValidationSplitPercentage",
default_value=0.05,
)
test_split_percentage = ParameterFloat(
name="TestSplitPercentage",
default_value=0.05,
)
feature_store_offline_prefix = ParameterString(
name="FeatureStoreOfflinePrefix",
default_value="reviews-feature-store-" + str(timestamp),
)
feature_group_name = ParameterString(
name="FeatureGroupName",
default_value="reviews-feature-group-" + str(timestamp))
train_instance_type = ParameterString(name="TrainInstanceType",
default_value="ml.c5.9xlarge")
train_instance_count = ParameterInteger(name="TrainInstanceCount",
default_value=1)
#########################
# PROCESSING STEP
#########################
processor = SKLearnProcessor(
framework_version="0.23-1",
role=role,
instance_type=processing_instance_type,
instance_count=processing_instance_count,
env={"AWS_DEFAULT_REGION": region},
max_runtime_in_seconds=7200,
)
processing_inputs = [
ProcessingInput(
input_name="raw-input-data",
source=input_data,
destination="/opt/ml/processing/input/data/",
s3_data_distribution_type="ShardedByS3Key",
)
]
processing_outputs = [
ProcessingOutput(
output_name="bert-train",
s3_upload_mode="EndOfJob",
source="/opt/ml/processing/output/bert/train",
),
ProcessingOutput(
output_name="bert-validation",
s3_upload_mode="EndOfJob",
source="/opt/ml/processing/output/bert/validation",
),
ProcessingOutput(
output_name="bert-test",
s3_upload_mode="EndOfJob",
source="/opt/ml/processing/output/bert/test",
),
]
# TODO: Figure out why the Parameter's are not resolving properly to their native type when user here.
# We shouldn't be using `default_value`
processing_step = ProcessingStep(
name="Processing",
processor=processor,
inputs=processing_inputs,
outputs=processing_outputs,
job_arguments=[
"--train-split-percentage",
str(train_split_percentage.default_value),
"--validation-split-percentage",
str(validation_split_percentage.default_value),
"--test-split-percentage",
str(test_split_percentage.default_value),
"--max-seq-length",
str(max_seq_length.default_value),
"--balance-dataset",
str(balance_dataset.default_value),
"--feature-store-offline-prefix",
str(feature_store_offline_prefix.default_value),
"--feature-group-name",
str(feature_group_name.default_value),
],
code=os.path.join(BASE_DIR,
"preprocess-scikit-text-to-bert-feature-store.py"),
)
#########################
# TRAINING STEP
#########################
epochs = ParameterInteger(name="Epochs", default_value=1)
learning_rate = ParameterFloat(name="LearningRate", default_value=0.00001)
epsilon = ParameterFloat(name="Epsilon", default_value=0.00000001)
train_batch_size = ParameterInteger(name="TrainBatchSize",
default_value=128)
validation_batch_size = ParameterInteger(name="ValidationBatchSize",
default_value=128)
test_batch_size = ParameterInteger(name="TestBatchSize", default_value=128)
train_steps_per_epoch = ParameterInteger(name="TrainStepsPerEpoch",
default_value=50)
validation_steps = ParameterInteger(name="ValidationSteps",
default_value=50)
test_steps = ParameterInteger(name="TestSteps", default_value=50)
train_volume_size = ParameterInteger(name="TrainVolumeSize",
default_value=1024)
use_xla = ParameterString(
name="UseXLA",
default_value="True",
)
use_amp = ParameterString(
name="UseAMP",
default_value="True",
)
freeze_bert_layer = ParameterString(
name="FreezeBERTLayer",
default_value="False",
)
enable_sagemaker_debugger = ParameterString(
name="EnableSageMakerDebugger",
default_value="False",
)
enable_checkpointing = ParameterString(
name="EnableCheckpointing",
default_value="False",
)
enable_tensorboard = ParameterString(
name="EnableTensorboard",
default_value="False",
)
input_mode = ParameterString(
name="InputMode",
default_value="File",
)
run_validation = ParameterString(
name="RunValidation",
default_value="True",
)
run_test = ParameterString(
name="RunTest",
default_value="False",
)
run_sample_predictions = ParameterString(
name="RunSamplePredictions",
default_value="False",
)
metrics_definitions = [
{
"Name": "train:loss",
"Regex": "loss: ([0-9\\.]+)"
},
{
"Name": "train:accuracy",
"Regex": "accuracy: ([0-9\\.]+)"
},
{
"Name": "validation:loss",
"Regex": "val_loss: ([0-9\\.]+)"
},
{
"Name": "validation:accuracy",
"Regex": "val_accuracy: ([0-9\\.]+)"
},
]
train_src = os.path.join(BASE_DIR, "src")
model_path = f"s3://{default_bucket}/{base_job_prefix}/output/model"
estimator = TensorFlow(
entry_point="tf_bert_reviews.py",
source_dir=BASE_DIR,
role=role,
output_path=model_path,
instance_count=train_instance_count,
instance_type=train_instance_type,
volume_size=train_volume_size,
py_version="py37",
framework_version="2.3.1",
hyperparameters={
"epochs": epochs,
"learning_rate": learning_rate,
"epsilon": epsilon,
"train_batch_size": train_batch_size,
"validation_batch_size": validation_batch_size,
"test_batch_size": test_batch_size,
"train_steps_per_epoch": train_steps_per_epoch,
"validation_steps": validation_steps,
"test_steps": test_steps,
"use_xla": use_xla,
"use_amp": use_amp,
"max_seq_length": max_seq_length,
"freeze_bert_layer": freeze_bert_layer,
"enable_sagemaker_debugger": enable_sagemaker_debugger,
"enable_checkpointing": enable_checkpointing,
"enable_tensorboard": enable_tensorboard,
"run_validation": run_validation,
"run_test": run_test,
"run_sample_predictions": run_sample_predictions,
},
input_mode=input_mode,
metric_definitions=metrics_definitions,
# max_run=7200 # max 2 hours * 60 minutes seconds per hour * 60 seconds per minute
)
training_step = TrainingStep(
name="Train",
estimator=estimator,
inputs={
"train":
TrainingInput(
s3_data=processing_step.properties.ProcessingOutputConfig.
Outputs["bert-train"].S3Output.S3Uri,
content_type="text/csv",
),
"validation":
TrainingInput(
s3_data=processing_step.properties.ProcessingOutputConfig.
Outputs["bert-validation"].S3Output.S3Uri,
content_type="text/csv",
),
"test":
TrainingInput(
s3_data=processing_step.properties.ProcessingOutputConfig.
Outputs["bert-test"].S3Output.S3Uri,
content_type="text/csv",
),
},
)
#########################
# EVALUATION STEP
#########################
evaluation_processor = SKLearnProcessor(
framework_version="0.23-1",
role=role,
instance_type=processing_instance_type,
instance_count=processing_instance_count,
env={"AWS_DEFAULT_REGION": region},
max_runtime_in_seconds=7200,
)
evaluation_report = PropertyFile(name="EvaluationReport",
output_name="metrics",
path="evaluation.json")
evaluation_step = ProcessingStep(
name="EvaluateModel",
processor=evaluation_processor,
code=os.path.join(BASE_DIR, "evaluate_model_metrics.py"),
inputs=[
ProcessingInput(
source=training_step.properties.ModelArtifacts.
S3ModelArtifacts,
destination="/opt/ml/processing/input/model",
),
ProcessingInput(
source=processing_step.properties.
ProcessingInputs["raw-input-data"].S3Input.S3Uri,
destination="/opt/ml/processing/input/data",
),
],
outputs=[
ProcessingOutput(output_name="metrics",
s3_upload_mode="EndOfJob",
source="/opt/ml/processing/output/metrics/"),
],
job_arguments=[
"--max-seq-length",
str(max_seq_length.default_value),
],
property_files=[evaluation_report
], # these cause deserialization issues
)
model_metrics = ModelMetrics(model_statistics=MetricsSource(
s3_uri="{}/evaluation.json".format(
evaluation_step.arguments["ProcessingOutputConfig"]["Outputs"][0]
["S3Output"]["S3Uri"]),
content_type="application/json",
))
#########################
## REGISTER TRAINED MODEL STEP
#########################
model_approval_status = ParameterString(
name="ModelApprovalStatus", default_value="PendingManualApproval")
deploy_instance_type = ParameterString(name="DeployInstanceType",
default_value="ml.m5.4xlarge")
deploy_instance_count = ParameterInteger(name="DeployInstanceCount",
default_value=1)
inference_image_uri = sagemaker.image_uris.retrieve(
framework="tensorflow",
region=region,
version="2.3.1",
py_version="py37",
instance_type=deploy_instance_type,
image_scope="inference",
)
print(inference_image_uri)
register_step = RegisterModel(
name="RegisterModel",
estimator=estimator,
image_uri=
inference_image_uri, # we have to specify, by default it's using training image
model_data=training_step.properties.ModelArtifacts.S3ModelArtifacts,
content_types=["text/csv"],
response_types=["text/csv"],
inference_instances=[
deploy_instance_type
], # The JSON spec must be within these instance types or we will see "Instance Type Not Allowed" Exception
transform_instances=[deploy_instance_type],
model_package_group_name=model_package_group_name,
approval_status=model_approval_status,
)
#########################
## CREATE MODEL FOR DEPLOYMENT STEP
#########################
model = Model(
image_uri=inference_image_uri,
model_data=training_step.properties.ModelArtifacts.S3ModelArtifacts,
sagemaker_session=sess,
role=role,
)
create_inputs = CreateModelInput(instance_type=deploy_instance_type, )
create_step = CreateModelStep(
name="CreateModel",
model=model,
inputs=create_inputs,
)
#########################
## CONDITION STEP: EVALUATE THE MODEL
#########################
min_accuracy_value = ParameterFloat(name="MinAccuracyValue",
default_value=0.01)
minimum_accuracy_condition = ConditionGreaterThanOrEqualTo(
left=JsonGet(
step=evaluation_step,
property_file=evaluation_report,
json_path="metrics.accuracy.value",
),
right=min_accuracy_value, # accuracy
)
minimum_accuracy_condition_step = ConditionStep(
name="AccuracyCondition",
conditions=[minimum_accuracy_condition],
if_steps=[register_step,
create_step], # success, continue with model registration
else_steps=[], # fail, end the pipeline
)
#########################
## CREATE PIPELINE
#########################
pipeline = Pipeline(
name=pipeline_name,
parameters=[
input_data,
processing_instance_count,
processing_instance_type,
max_seq_length,
balance_dataset,
train_split_percentage,
validation_split_percentage,
test_split_percentage,
feature_store_offline_prefix,
feature_group_name,
train_instance_type,
train_instance_count,
epochs,
learning_rate,
epsilon,
train_batch_size,
validation_batch_size,
test_batch_size,
train_steps_per_epoch,
validation_steps,
test_steps,
train_volume_size,
use_xla,
use_amp,
freeze_bert_layer,
enable_sagemaker_debugger,
enable_checkpointing,
enable_tensorboard,
input_mode,
run_validation,
run_test,
run_sample_predictions,
min_accuracy_value,
model_approval_status,
deploy_instance_type,
deploy_instance_count,
],
steps=[
processing_step, training_step, evaluation_step,
minimum_accuracy_condition_step
],
sagemaker_session=sess,
)
#########################
## RETURN PIPELINE
#########################
return pipeline
def test_estimator_transformer(estimator):
model_data = f"s3://{BUCKET}/model.tar.gz"
model_inputs = CreateModelInput(
instance_type="c4.4xlarge",
accelerator_type="ml.eia1.medium",
)
service_fault_retry_policy = StepRetryPolicy(
exception_types=[StepExceptionTypeEnum.SERVICE_FAULT], max_attempts=10)
transform_inputs = TransformInput(data=f"s3://{BUCKET}/transform_manifest")
estimator_transformer = EstimatorTransformer(
name="EstimatorTransformerStep",
estimator=estimator,
model_data=model_data,
model_inputs=model_inputs,
instance_count=1,
instance_type="ml.c4.4xlarge",
transform_inputs=transform_inputs,
depends_on=["TestStep"],
model_step_retry_policies=[service_fault_retry_policy],
transform_step_retry_policies=[service_fault_retry_policy],
repack_model_step_retry_policies=[service_fault_retry_policy],
)
request_dicts = estimator_transformer.request_dicts()
assert len(request_dicts) == 2
for request_dict in request_dicts:
if request_dict["Type"] == "Model":
assert request_dict == {
"Name": "EstimatorTransformerStepCreateModelStep",
"Type": "Model",
"DependsOn": ["TestStep"],
"RetryPolicies": [service_fault_retry_policy.to_request()],
"Arguments": {
"ExecutionRoleArn": "DummyRole",
"PrimaryContainer": {
"Environment": {},
"Image": "fakeimage",
"ModelDataUrl": "s3://my-bucket/model.tar.gz",
},
},
}
elif request_dict["Type"] == "Transform":
assert request_dict[
"Name"] == "EstimatorTransformerStepTransformStep"
assert request_dict["RetryPolicies"] == [
service_fault_retry_policy.to_request()
]
arguments = request_dict["Arguments"]
assert isinstance(arguments["ModelName"], Properties)
arguments.pop("ModelName")
assert "DependsOn" not in request_dict
assert arguments == {
"TransformInput": {
"DataSource": {
"S3DataSource": {
"S3DataType": "S3Prefix",
"S3Uri": f"s3://{BUCKET}/transform_manifest",
}
}
},
"TransformOutput": {
"S3OutputPath": None
},
"TransformResources": {
"InstanceCount": 1,
"InstanceType": "ml.c4.4xlarge"
},
}
else:
raise Exception(
"A step exists in the collection of an invalid type.")
def test_end_to_end_pipeline_successful_execution(
sagemaker_session, region_name, role, pipeline_name, wait=False
):
model_package_group_name = f"{pipeline_name}ModelPackageGroup"
data_path = os.path.join(DATA_DIR, "workflow")
default_bucket = sagemaker_session.default_bucket()
# download the input data
local_input_path = os.path.join(data_path, "abalone-dataset.csv")
s3 = sagemaker_session.boto_session.resource("s3")
s3.Bucket(f"sagemaker-servicecatalog-seedcode-{region_name}").download_file(
"dataset/abalone-dataset.csv", local_input_path
)
# # upload the input data to our bucket
base_uri = f"s3://{default_bucket}/{pipeline_name}"
with open(local_input_path) as data:
body = data.read()
input_data_uri = S3Uploader.upload_string_as_file_body(
body=body,
desired_s3_uri=f"{base_uri}/abalone-dataset.csv",
sagemaker_session=sagemaker_session,
)
# download batch transform data
local_batch_path = os.path.join(data_path, "abalone-dataset-batch")
s3.Bucket(f"sagemaker-servicecatalog-seedcode-{region_name}").download_file(
"dataset/abalone-dataset-batch", local_batch_path
)
# upload the batch transform data
with open(local_batch_path) as data:
body = data.read()
batch_data_uri = S3Uploader.upload_string_as_file_body(
body=body,
desired_s3_uri=f"{base_uri}/abalone-dataset-batch",
sagemaker_session=sagemaker_session,
)
# define parameters
processing_instance_count = ParameterInteger(name="ProcessingInstanceCount", default_value=1)
processing_instance_type = ParameterString(
name="ProcessingInstanceType", default_value="ml.m5.xlarge"
)
training_instance_type = ParameterString(
name="TrainingInstanceType", default_value="ml.m5.xlarge"
)
model_approval_status = ParameterString(name="ModelApprovalStatus", default_value="Approved")
input_data = ParameterString(
name="InputData",
default_value=input_data_uri,
)
batch_data = ParameterString(
name="BatchData",
default_value=batch_data_uri,
)
# define processing step
framework_version = "0.23-1"
sklearn_processor = SKLearnProcessor(
framework_version=framework_version,
instance_type=processing_instance_type,
instance_count=processing_instance_count,
base_job_name=f"{pipeline_name}-process",
role=role,
sagemaker_session=sagemaker_session,
)
step_process = ProcessingStep(
name="AbaloneProcess",
processor=sklearn_processor,
inputs=[
ProcessingInput(source=input_data, destination="/opt/ml/processing/input"),
],
outputs=[
ProcessingOutput(output_name="train", source="/opt/ml/processing/train"),
ProcessingOutput(output_name="validation", source="/opt/ml/processing/validation"),
ProcessingOutput(output_name="test", source="/opt/ml/processing/test"),
],
code=os.path.join(data_path, "abalone/preprocessing.py"),
)
# define training step
model_path = f"s3://{default_bucket}/{pipeline_name}Train"
image_uri = image_uris.retrieve(
framework="xgboost",
region=region_name,
version="1.0-1",
py_version="py3",
instance_type=training_instance_type,
)
xgb_train = Estimator(
image_uri=image_uri,
instance_type=training_instance_type,
instance_count=1,
output_path=model_path,
role=role,
sagemaker_session=sagemaker_session,
)
xgb_train.set_hyperparameters(
objective="reg:linear",
num_round=50,
max_depth=5,
eta=0.2,
gamma=4,
min_child_weight=6,
subsample=0.7,
silent=0,
)
step_train = TrainingStep(
name="AbaloneTrain",
estimator=xgb_train,
inputs={
"train": TrainingInput(
s3_data=step_process.properties.ProcessingOutputConfig.Outputs[
"train"
].S3Output.S3Uri,
content_type="text/csv",
),
"validation": TrainingInput(
s3_data=step_process.properties.ProcessingOutputConfig.Outputs[
"validation"
].S3Output.S3Uri,
content_type="text/csv",
),
},
)
# define evaluation step
script_eval = ScriptProcessor(
image_uri=image_uri,
command=["python3"],
instance_type=processing_instance_type,
instance_count=1,
base_job_name=f"{pipeline_name}-eval",
role=role,
sagemaker_session=sagemaker_session,
)
evaluation_report = PropertyFile(
name="EvaluationReport", output_name="evaluation", path="evaluation.json"
)
step_eval = ProcessingStep(
name="AbaloneEval",
processor=script_eval,
inputs=[
ProcessingInput(
source=step_train.properties.ModelArtifacts.S3ModelArtifacts,
destination="/opt/ml/processing/model",
),
ProcessingInput(
source=step_process.properties.ProcessingOutputConfig.Outputs[
"test"
].S3Output.S3Uri,
destination="/opt/ml/processing/test",
),
],
outputs=[
ProcessingOutput(output_name="evaluation", source="/opt/ml/processing/evaluation"),
],
code=os.path.join(data_path, "abalone/evaluation.py"),
property_files=[evaluation_report],
)
# define create model step
model = Model(
image_uri=image_uri,
model_data=step_train.properties.ModelArtifacts.S3ModelArtifacts,
sagemaker_session=sagemaker_session,
role=role,
)
inputs = CreateModelInput(
instance_type="ml.m5.large",
accelerator_type="ml.eia1.medium",
)
step_create_model = CreateModelStep(
name="AbaloneCreateModel",
model=model,
inputs=inputs,
)
# define transform step
transformer = Transformer(
model_name=step_create_model.properties.ModelName,
instance_type="ml.m5.xlarge",
instance_count=1,
output_path=f"s3://{default_bucket}/{pipeline_name}Transform",
sagemaker_session=sagemaker_session,
)
step_transform = TransformStep(
name="AbaloneTransform",
transformer=transformer,
inputs=TransformInput(data=batch_data),
)
# define register model step
model_metrics = ModelMetrics(
model_statistics=MetricsSource(
s3_uri="{}/evaluation.json".format(
step_eval.arguments["ProcessingOutputConfig"]["Outputs"][0]["S3Output"]["S3Uri"]
),
content_type="application/json",
)
)
step_register = RegisterModel(
name="AbaloneRegisterModel",
estimator=xgb_train,
model_data=step_train.properties.ModelArtifacts.S3ModelArtifacts,
content_types=["text/csv"],
response_types=["text/csv"],
inference_instances=["ml.t2.medium", "ml.m5.xlarge"],
transform_instances=["ml.m5.xlarge"],
model_package_group_name=model_package_group_name,
approval_status=model_approval_status,
model_metrics=model_metrics,
)
# define condition step
cond_lte = ConditionLessThanOrEqualTo(
left=JsonGet(
step_name=step_eval.name,
property_file=evaluation_report,
json_path="regression_metrics.mse.value",
),
right=20.0,
)
step_cond = ConditionStep(
name="AbaloneMSECond",
conditions=[cond_lte],
if_steps=[step_register, step_create_model, step_transform],
else_steps=[],
)
# define pipeline
pipeline = Pipeline(
name=pipeline_name,
parameters=[
processing_instance_type,
processing_instance_count,
training_instance_type,
model_approval_status,
input_data,
batch_data,
],
steps=[step_process, step_train, step_eval, step_cond],
sagemaker_session=sagemaker_session,
)
pipeline.create(role)
execution = pipeline.start()
execution_arn = execution.arn
if wait:
execution.wait()
return execution_arn
def test_steps_with_map_params_pipeline(
sagemaker_session,
role,
script_dir,
pipeline_name,
region_name,
athena_dataset_definition,
):
instance_count = ParameterInteger(name="InstanceCount", default_value=2)
framework_version = "0.20.0"
instance_type = ParameterString(name="InstanceType", default_value="ml.m5.xlarge")
output_prefix = ParameterString(name="OutputPrefix", default_value="output")
input_data = f"s3://sagemaker-sample-data-{region_name}/processing/census/census-income.csv"
sklearn_processor = SKLearnProcessor(
framework_version=framework_version,
instance_type=instance_type,
instance_count=instance_count,
base_job_name="test-sklearn",
sagemaker_session=sagemaker_session,
role=role,
)
step_process = ProcessingStep(
name="my-process",
display_name="ProcessingStep",
description="description for Processing step",
processor=sklearn_processor,
inputs=[
ProcessingInput(source=input_data, destination="/opt/ml/processing/input"),
ProcessingInput(dataset_definition=athena_dataset_definition),
],
outputs=[
ProcessingOutput(output_name="train_data", source="/opt/ml/processing/train"),
ProcessingOutput(
output_name="test_data",
source="/opt/ml/processing/test",
destination=Join(
on="/",
values=[
"s3:/",
sagemaker_session.default_bucket(),
"test-sklearn",
output_prefix,
ExecutionVariables.PIPELINE_EXECUTION_ID,
],
),
),
],
code=os.path.join(script_dir, "preprocessing.py"),
)
sklearn_train = SKLearn(
framework_version=framework_version,
entry_point=os.path.join(script_dir, "train.py"),
instance_type=instance_type,
sagemaker_session=sagemaker_session,
role=role,
hyperparameters={
"batch-size": 500,
"epochs": 5,
},
)
step_train = TrainingStep(
name="my-train",
display_name="TrainingStep",
description="description for Training step",
estimator=sklearn_train,
inputs=TrainingInput(
s3_data=step_process.properties.ProcessingOutputConfig.Outputs[
"train_data"
].S3Output.S3Uri
),
)
model = Model(
image_uri=sklearn_train.image_uri,
model_data=step_train.properties.ModelArtifacts.S3ModelArtifacts,
sagemaker_session=sagemaker_session,
role=role,
)
model_inputs = CreateModelInput(
instance_type="ml.m5.large",
accelerator_type="ml.eia1.medium",
)
step_model = CreateModelStep(
name="my-model",
display_name="ModelStep",
description="description for Model step",
model=model,
inputs=model_inputs,
)
# Condition step for evaluating model quality and branching execution
cond_lte = ConditionGreaterThanOrEqualTo(
left=step_train.properties.HyperParameters["batch-size"],
right=6.0,
)
step_cond = ConditionStep(
name="CustomerChurnAccuracyCond",
conditions=[cond_lte],
if_steps=[],
else_steps=[step_model],
)
pipeline = Pipeline(
name=pipeline_name,
parameters=[instance_type, instance_count, output_prefix],
steps=[step_process, step_train, step_cond],
sagemaker_session=sagemaker_session,
)
definition = json.loads(pipeline.definition())
assert definition["Version"] == "2020-12-01"
steps = definition["Steps"]
assert len(steps) == 3
training_args = {}
condition_args = {}
for step in steps:
if step["Type"] == "Training":
training_args = step["Arguments"]
if step["Type"] == "Condition":
condition_args = step["Arguments"]
assert training_args["InputDataConfig"][0]["DataSource"]["S3DataSource"]["S3Uri"] == {
"Get": "Steps.my-process.ProcessingOutputConfig.Outputs['train_data'].S3Output.S3Uri"
}
assert condition_args["Conditions"][0]["LeftValue"] == {
"Get": "Steps.my-train.HyperParameters['batch-size']"
}
try:
response = pipeline.create(role)
create_arn = response["PipelineArn"]
assert re.match(
rf"arn:aws:sagemaker:{region_name}:\d{{12}}:pipeline/{pipeline_name}",
create_arn,
)
finally:
try:
pipeline.delete()
except Exception:
pass
def test_tuning_single_algo(
sagemaker_session,
role,
cpu_instance_type,
pipeline_name,
region_name,
):
base_dir = os.path.join(DATA_DIR, "pytorch_mnist")
entry_point = os.path.join(base_dir, "mnist.py")
input_path = sagemaker_session.upload_data(
path=os.path.join(base_dir, "training"),
key_prefix="integ-test-data/pytorch_mnist/training",
)
inputs = TrainingInput(s3_data=input_path)
instance_count = ParameterInteger(name="InstanceCount", default_value=1)
instance_type = ParameterString(name="InstanceType",
default_value="ml.m5.xlarge")
pytorch_estimator = PyTorch(
entry_point=entry_point,
role=role,
framework_version="1.5.0",
py_version="py3",
instance_count=instance_count,
instance_type=instance_type,
sagemaker_session=sagemaker_session,
enable_sagemaker_metrics=True,
max_retry_attempts=3,
)
min_batch_size = ParameterInteger(name="MinBatchSize", default_value=64)
max_batch_size = ParameterInteger(name="MaxBatchSize", default_value=128)
hyperparameter_ranges = {
"batch-size": IntegerParameter(min_batch_size, max_batch_size),
}
tuner = HyperparameterTuner(
estimator=pytorch_estimator,
objective_metric_name="test:acc",
objective_type="Maximize",
hyperparameter_ranges=hyperparameter_ranges,
metric_definitions=[{
"Name": "test:acc",
"Regex": "Overall test accuracy: (.*?);"
}],
max_jobs=2,
max_parallel_jobs=2,
)
step_tune = TuningStep(
name="my-tuning-step",
tuner=tuner,
inputs=inputs,
)
best_model = Model(
image_uri=pytorch_estimator.training_image_uri(),
model_data=step_tune.get_top_model_s3_uri(
top_k=0,
s3_bucket=sagemaker_session.default_bucket(),
),
sagemaker_session=sagemaker_session,
role=role,
)
model_inputs = CreateModelInput(
instance_type="ml.m5.large",
accelerator_type="ml.eia1.medium",
)
step_best_model = CreateModelStep(
name="1st-model",
model=best_model,
inputs=model_inputs,
)
second_best_model = Model(
image_uri=pytorch_estimator.training_image_uri(),
model_data=step_tune.get_top_model_s3_uri(
top_k=1,
s3_bucket=sagemaker_session.default_bucket(),
),
sagemaker_session=sagemaker_session,
role=role,
entry_point=entry_point,
source_dir=base_dir,
)
step_second_best_model = CreateModelStep(
name="2nd-best-model",
model=second_best_model,
inputs=model_inputs,
)
pipeline = Pipeline(
name=pipeline_name,
parameters=[
instance_count, instance_type, min_batch_size, max_batch_size
],
steps=[step_tune, step_best_model, step_second_best_model],
sagemaker_session=sagemaker_session,
)
try:
response = pipeline.create(role)
create_arn = response["PipelineArn"]
assert re.match(
rf"arn:aws:sagemaker:{region_name}:\d{{12}}:pipeline/{pipeline_name}",
create_arn,
)
for _ in retries(
max_retry_count=5,
exception_message_prefix=
"Waiting for a successful execution of pipeline",
seconds_to_sleep=10,
):
execution = pipeline.start(parameters={})
assert re.match(
rf"arn:aws:sagemaker:{region_name}:\d{{12}}:pipeline/{pipeline_name}/execution/",
execution.arn,
)
try:
execution.wait(delay=30, max_attempts=60)
except WaiterError:
pass
execution_steps = execution.list_steps()
assert len(execution_steps) == 3
for step in execution_steps:
assert step["StepStatus"] == "Succeeded"
break
finally:
try:
pipeline.delete()
except Exception:
pass
def test_estimator_transformer_with_model_repack_with_estimator(estimator):
model_data = f"s3://{BUCKET}/model.tar.gz"
model_inputs = CreateModelInput(
instance_type="c4.4xlarge",
accelerator_type="ml.eia1.medium",
)
service_fault_retry_policy = StepRetryPolicy(
exception_types=[StepExceptionTypeEnum.SERVICE_FAULT], max_attempts=10
)
transform_inputs = TransformInput(data=f"s3://{BUCKET}/transform_manifest")
estimator_transformer = EstimatorTransformer(
name="EstimatorTransformerStep",
estimator=estimator,
model_data=model_data,
model_inputs=model_inputs,
instance_count=1,
instance_type="ml.c4.4xlarge",
transform_inputs=transform_inputs,
depends_on=["TestStep"],
model_step_retry_policies=[service_fault_retry_policy],
transform_step_retry_policies=[service_fault_retry_policy],
repack_model_step_retry_policies=[service_fault_retry_policy],
entry_point=f"{DATA_DIR}/dummy_script.py",
)
request_dicts = estimator_transformer.request_dicts()
assert len(request_dicts) == 3
for request_dict in request_dicts:
if request_dict["Type"] == "Training":
assert request_dict["Name"] == "EstimatorTransformerStepRepackModel"
assert request_dict["DependsOn"] == ["TestStep"]
assert request_dict["RetryPolicies"] == [service_fault_retry_policy.to_request()]
arguments = request_dict["Arguments"]
# pop out the dynamic generated fields
arguments["HyperParameters"].pop("sagemaker_submit_directory")
assert arguments == {
"AlgorithmSpecification": {
"TrainingInputMode": "File",
"TrainingImage": "246618743249.dkr.ecr.us-west-2.amazonaws.com/"
+ "sagemaker-scikit-learn:0.23-1-cpu-py3",
},
"OutputDataConfig": {"S3OutputPath": "s3://my-bucket/"},
"StoppingCondition": {"MaxRuntimeInSeconds": 86400},
"ResourceConfig": {
"InstanceCount": 1,
"InstanceType": "ml.m5.large",
"VolumeSizeInGB": 30,
},
"RoleArn": "DummyRole",
"InputDataConfig": [
{
"DataSource": {
"S3DataSource": {
"S3DataType": "S3Prefix",
"S3Uri": "s3://my-bucket/model.tar.gz",
"S3DataDistributionType": "FullyReplicated",
}
},
"ChannelName": "training",
}
],
"HyperParameters": {
"inference_script": '"dummy_script.py"',
"model_archive": '"s3://my-bucket/model.tar.gz"',
"dependencies": "null",
"source_dir": "null",
"sagemaker_program": '"_repack_model.py"',
"sagemaker_container_log_level": "20",
"sagemaker_region": '"us-west-2"',
},
"VpcConfig": {"Subnets": ["abc", "def"], "SecurityGroupIds": ["123", "456"]},
"DebugHookConfig": {
"S3OutputPath": "s3://my-bucket/",
"CollectionConfigurations": [],
},
}
elif request_dict["Type"] == "Model":
assert request_dict["Name"] == "EstimatorTransformerStepCreateModelStep"
assert request_dict["RetryPolicies"] == [service_fault_retry_policy.to_request()]
arguments = request_dict["Arguments"]
assert isinstance(arguments["PrimaryContainer"]["ModelDataUrl"], Properties)
arguments["PrimaryContainer"].pop("ModelDataUrl")
assert "DependsOn" not in request_dict
assert arguments == {
"ExecutionRoleArn": "DummyRole",
"PrimaryContainer": {
"Environment": {},
"Image": "fakeimage",
},
}
elif request_dict["Type"] == "Transform":
assert request_dict["Name"] == "EstimatorTransformerStepTransformStep"
assert request_dict["RetryPolicies"] == [service_fault_retry_policy.to_request()]
arguments = request_dict["Arguments"]
assert isinstance(arguments["ModelName"], Properties)
arguments.pop("ModelName")
assert "DependsOn" not in request_dict
assert arguments == {
"TransformInput": {
"DataSource": {
"S3DataSource": {
"S3DataType": "S3Prefix",
"S3Uri": f"s3://{BUCKET}/transform_manifest",
}
}
},
"TransformOutput": {"S3OutputPath": None},
"TransformResources": {"InstanceCount": 1, "InstanceType": "ml.c4.4xlarge"},
}
else:
raise Exception("A step exists in the collection of an invalid type.")