def test_auto_ml_attach(sagemaker_session):
expected_default_input_config = [{
"DataSource": {
"S3DataSource": {
"S3DataType":
"S3Prefix",
"S3Uri":
"s3://{}/{}/input/iris_training.csv".format(
sagemaker_session.default_bucket(), PREFIX),
}
},
"TargetAttributeName":
TARGET_ATTRIBUTE_NAME,
}]
expected_default_output_config = {
"S3OutputPath": "s3://{}/".format(sagemaker_session.default_bucket())
}
auto_ml_utils.create_auto_ml_job_if_not_exist(sagemaker_session)
attached_automl_job = AutoML.attach(auto_ml_job_name=AUTO_ML_JOB_NAME,
sagemaker_session=sagemaker_session)
attached_desc = attached_automl_job.describe_auto_ml_job()
assert attached_desc["AutoMLJobName"] == AUTO_ML_JOB_NAME
assert attached_desc["AutoMLJobStatus"] == "Completed"
assert isinstance(attached_desc["BestCandidate"], dict)
assert attached_desc["InputDataConfig"] == expected_default_input_config
assert attached_desc["AutoMLJobConfig"] == EXPECTED_DEFAULT_JOB_CONFIG
assert attached_desc["OutputDataConfig"] == expected_default_output_config
def test_create_model_best_candidate(sagemaker_session, cpu_instance_type):
auto_ml_utils.create_auto_ml_job_if_not_exist(sagemaker_session)
auto_ml = AutoML.attach(auto_ml_job_name=AUTO_ML_JOB_NAME, sagemaker_session=sagemaker_session)
best_candidate = auto_ml.best_candidate()
with timeout(minutes=5):
pipeline_model = auto_ml.create_model(
name=DEFAULT_MODEL_NAME,
candidate=best_candidate,
sagemaker_session=sagemaker_session,
vpc_config=None,
enable_network_isolation=False,
model_kms_key=None,
predictor_cls=None,
)
inputs = sagemaker_session.upload_data(
path=TRANSFORM_DATA, key_prefix=PREFIX + "/transform_input"
)
pipeline_model.transformer(
instance_count=1,
instance_type=cpu_instance_type,
assemble_with="Line",
output_path="s3://{}/{}".format(sagemaker_session.default_bucket(), "transform_test"),
accept="text/csv",
).transform(data=inputs, content_type="text/csv", split_type="Line", join_source="Input")
def test_attach(sagemaker_session):
aml = AutoML.attach(auto_ml_job_name=JOB_NAME_3, sagemaker_session=sagemaker_session)
assert aml.current_job_name == JOB_NAME_3
assert aml.role == "mock_role_arn"
assert aml.target_attribute_name == "y"
assert aml.problem_type == "Auto"
assert aml.output_path == "s3://output_prefix"
assert aml.tags == LIST_TAGS_RESULT["Tags"]
def get_candidates(top_n_candidates, job_name):
# takes an autopilot job name, returns top candidates
est = AutoML.attach(auto_ml_job_name = job_name)
candidates = est.list_candidates(sort_by='FinalObjectiveMetricValue',
sort_order='Descending',
max_results=top_n_candidates)
return est, candidates
def lambda_handler(event, context):
try :
model_config = event["Input"]["taskresult"]["Payload"]["model-config"]
automl_config = event["Input"]["taskresult"]["Payload"]["automl-config"]
security_config = event["Input"]["taskresult"]["Payload"]["security-config"]
session = Session()
automl_job = AutoML.attach(automl_config["job_name"],
sagemaker_session=session)
model = automl_job.create_model(model_config["model_name"],
inference_response_keys=model_config["inference_response_keys"])
model.models[0].env["AUTOML_SPARSE_ENCODE_RECORDIO_PROTOBUF"] = "1"
session.create_model(name=model_config["model_name"],
role = security_config["iam_role"],
container_defs= model.pipeline_container_def(model_config["instance_type"]))
except KeyError as e:
raise KeyError(f"KeyError on input: {event}")
return event["Input"]["taskresult"]["Payload"]
