def test_auto_ml_fit_optional_args(sagemaker_session):
output_path = "s3://{}/{}".format(sagemaker_session.default_bucket(),
"specified_ouput_path")
problem_type = "MulticlassClassification"
job_objective = {"MetricName": "Accuracy"}
auto_ml = AutoML(
role=ROLE,
target_attribute_name=TARGET_ATTRIBUTE_NAME,
sagemaker_session=sagemaker_session,
max_candidates=1,
output_path=output_path,
problem_type=problem_type,
job_objective=job_objective,
generate_candidate_definitions_only=True,
)
inputs = TRAINING_DATA
with timeout(minutes=AUTO_ML_DEFAULT_TIMEMOUT_MINUTES):
auto_ml.fit(inputs, job_name=unique_name_from_base(BASE_JOB_NAME))
auto_ml_desc = auto_ml.describe_auto_ml_job(
job_name=auto_ml.latest_auto_ml_job.job_name)
assert auto_ml_desc["AutoMLJobStatus"] == "Completed"
assert auto_ml_desc["AutoMLJobName"] == auto_ml.latest_auto_ml_job.job_name
assert auto_ml_desc["AutoMLJobObjective"] == job_objective
assert auto_ml_desc["ProblemType"] == problem_type
assert auto_ml_desc["OutputDataConfig"]["S3OutputPath"] == output_path
def test_auto_ml_default_fit(strftime, sagemaker_session):
auto_ml = AutoML(
role=ROLE, target_attribute_name=TARGET_ATTRIBUTE_NAME, sagemaker_session=sagemaker_session
)
inputs = DEFAULT_S3_INPUT_DATA
auto_ml.fit(inputs)
sagemaker_session.auto_ml.assert_called_once()
_, args = sagemaker_session.auto_ml.call_args
assert args == {
"input_config": [
{
"DataSource": {
"S3DataSource": {"S3DataType": "S3Prefix", "S3Uri": DEFAULT_S3_INPUT_DATA}
},
"TargetAttributeName": TARGET_ATTRIBUTE_NAME,
}
],
"output_config": {"S3OutputPath": DEFAULT_OUTPUT_PATH},
"auto_ml_job_config": {
"CompletionCriteria": {"MaxCandidates": DEFAULT_MAX_CANDIDATES},
"SecurityConfig": {
"EnableInterContainerTrafficEncryption": ENCRYPT_INTER_CONTAINER_TRAFFIC
},
},
"role": ROLE,
"job_name": DEFAULT_JOB_NAME,
"problem_type": None,
"job_objective": None,
"generate_candidate_definitions_only": GENERATE_CANDIDATE_DEFINITIONS_ONLY,
"tags": None,
}
def test_auto_ml_fit_set_logs_to_false(start_new, sagemaker_session, caplog):
auto_ml = AutoML(
role=ROLE, target_attribute_name=TARGET_ATTRIBUTE_NAME, sagemaker_session=sagemaker_session
)
inputs = DEFAULT_S3_INPUT_DATA
auto_ml.fit(inputs, job_name=JOB_NAME, wait=False, logs=True)
start_new.wait.assert_not_called()
assert "Setting logs to False. logs is only meaningful when wait is True." in caplog.text
def test_auto_ml_local_input(sagemaker_session):
auto_ml = AutoML(
role=ROLE, target_attribute_name=TARGET_ATTRIBUTE_NAME, sagemaker_session=sagemaker_session
)
inputs = DEFAULT_S3_INPUT_DATA
auto_ml.fit(inputs)
sagemaker_session.auto_ml.assert_called_once()
_, args = sagemaker_session.auto_ml.call_args
assert args["input_config"][0]["DataSource"]["S3DataSource"]["S3Uri"] == DEFAULT_S3_INPUT_DATA
def test_auto_ml_additional_optional_params(sagemaker_session):
auto_ml = AutoML(
role=ROLE,
target_attribute_name=TARGET_ATTRIBUTE_NAME,
sagemaker_session=sagemaker_session,
volume_kms_key=VOLUME_KMS_KEY,
vpc_config=VPC_CONFIG,
encrypt_inter_container_traffic=ENCRYPT_INTER_CONTAINER_TRAFFIC,
compression_type=COMPRESSION_TYPE,
output_kms_key=OUTPUT_KMS_KEY,
output_path=OUTPUT_PATH,
problem_type=PROBLEM_TYPE,
max_candidates=MAX_CANDIDATES,
max_runtime_per_training_job_in_seconds=MAX_RUNTIME_PER_TRAINING_JOB,
total_job_runtime_in_seconds=TOTAL_JOB_RUNTIME,
job_objective=JOB_OBJECTIVE,
generate_candidate_definitions_only=GENERATE_CANDIDATE_DEFINITIONS_ONLY,
tags=TAGS,
)
inputs = DEFAULT_S3_INPUT_DATA
auto_ml.fit(inputs, job_name=JOB_NAME)
sagemaker_session.auto_ml.assert_called_once()
_, args = sagemaker_session.auto_ml.call_args
assert args == {
"input_config": [
{
"CompressionType": COMPRESSION_TYPE,
"DataSource": {
"S3DataSource": {"S3DataType": "S3Prefix", "S3Uri": DEFAULT_S3_INPUT_DATA}
},
"TargetAttributeName": TARGET_ATTRIBUTE_NAME,
}
],
"output_config": {"S3OutputPath": OUTPUT_PATH, "KmsKeyId": OUTPUT_KMS_KEY},
"auto_ml_job_config": {
"CompletionCriteria": {
"MaxAutoMLJobRuntimeInSeconds": TOTAL_JOB_RUNTIME,
"MaxCandidates": MAX_CANDIDATES,
"MaxRuntimePerTrainingJobInSeconds": MAX_RUNTIME_PER_TRAINING_JOB,
},
"SecurityConfig": {
"VolumeKmsKeyId": VOLUME_KMS_KEY,
"VpcConfig": VPC_CONFIG,
"EnableInterContainerTrafficEncryption": ENCRYPT_INTER_CONTAINER_TRAFFIC,
},
},
"job_name": JOB_NAME,
"role": ROLE,
"job_objective": JOB_OBJECTIVE,
"problem_type": PROBLEM_TYPE,
"generate_candidate_definitions_only": GENERATE_CANDIDATE_DEFINITIONS_ONLY,
"tags": TAGS,
}
def test_auto_ml_fit_local_input(sagemaker_session):
auto_ml = AutoML(
role=ROLE,
target_attribute_name=TARGET_ATTRIBUTE_NAME,
sagemaker_session=sagemaker_session,
max_candidates=1,
)
inputs = TRAINING_DATA
with timeout(minutes=AUTO_ML_DEFAULT_TIMEMOUT_MINUTES):
auto_ml.fit(inputs)
def test_auto_ml_invalid_target_attribute(sagemaker_session):
auto_ml = AutoML(role=ROLE,
target_attribute_name="y",
sagemaker_session=sagemaker_session,
max_candidates=1)
inputs = sagemaker_session.upload_data(path=TRAINING_DATA,
key_prefix=PREFIX + "/input")
with pytest.raises(
UnexpectedStatusException,
match="Could not complete the data builder processing job."):
auto_ml.fit(inputs)
def test_auto_ml_fit_local_input(sagemaker_session):
auto_ml = AutoML(
role=ROLE,
target_attribute_name=TARGET_ATTRIBUTE_NAME,
sagemaker_session=sagemaker_session,
max_candidates=1,
)
inputs = TRAINING_DATA
job_name = unique_name_from_base("auto-ml", max_length=32)
with timeout(minutes=AUTO_ML_DEFAULT_TIMEMOUT_MINUTES):
auto_ml.fit(inputs, job_name=job_name)
def test_auto_ml_input_object_fit(sagemaker_session):
auto_ml = AutoML(
role=ROLE,
target_attribute_name=TARGET_ATTRIBUTE_NAME,
sagemaker_session=sagemaker_session,
max_candidates=1,
)
job_name = unique_name_from_base("auto-ml", max_length=32)
s3_input = sagemaker_session.upload_data(path=TRAINING_DATA, key_prefix=PREFIX + "/input")
inputs = AutoMLInput(inputs=s3_input, target_attribute_name=TARGET_ATTRIBUTE_NAME)
with timeout(minutes=AUTO_ML_DEFAULT_TIMEMOUT_MINUTES):
auto_ml.fit(inputs, job_name=job_name)
def test_auto_ml_invalid_target_attribute(sagemaker_session):
auto_ml = AutoML(
role=ROLE, target_attribute_name="y", sagemaker_session=sagemaker_session, max_candidates=1
)
job_name = unique_name_from_base("auto-ml", max_length=32)
inputs = sagemaker_session.upload_data(path=TRAINING_DATA, key_prefix=PREFIX + "/input")
with pytest.raises(
ClientError,
match=r"An error occurred \(ValidationException\) when calling the CreateAutoMLJob "
"operation: Target attribute name y does not exist in header.",
):
auto_ml.fit(inputs, job_name=job_name)
def test_auto_ml_fit(sagemaker_session):
auto_ml = AutoML(
role=ROLE,
target_attribute_name=TARGET_ATTRIBUTE_NAME,
sagemaker_session=sagemaker_session,
max_candidates=3,
)
inputs = sagemaker_session.upload_data(path=TRAINING_DATA,
key_prefix=PREFIX + "/input")
with timeout(minutes=AUTO_ML_DEFAULT_TIMEMOUT_MINUTES):
auto_ml.fit(inputs)
def create_auto_ml_job_if_not_exist(sagemaker_session):
auto_ml_job_name = "python-sdk-integ-test-base-job"
try:
sagemaker_session.describe_auto_ml_job(job_name=auto_ml_job_name)
except Exception as e: # noqa: F841
auto_ml = AutoML(
role=ROLE,
target_attribute_name=TARGET_ATTRIBUTE_NAME,
sagemaker_session=sagemaker_session,
max_candidates=3,
)
inputs = sagemaker_session.upload_data(path=TRAINING_DATA, key_prefix=PREFIX + "/input")
with timeout(minutes=AUTO_ML_DEFAULT_TIMEMOUT_MINUTES):
auto_ml.fit(inputs, job_name=auto_ml_job_name, wait=True)
def test_auto_ml_input(sagemaker_session):
inputs = AutoMLInput(
inputs=DEFAULT_S3_INPUT_DATA, target_attribute_name="target", compression="Gzip"
)
auto_ml = AutoML(
role=ROLE, target_attribute_name=TARGET_ATTRIBUTE_NAME, sagemaker_session=sagemaker_session
)
auto_ml.fit(inputs)
_, args = sagemaker_session.auto_ml.call_args
assert args["input_config"] == [
{
"CompressionType": "Gzip",
"DataSource": {
"S3DataSource": {"S3DataType": "S3Prefix", "S3Uri": DEFAULT_S3_INPUT_DATA}
},
"TargetAttributeName": TARGET_ATTRIBUTE_NAME,
}
]
def test_auto_ml_fit_optional_args(sagemaker_session):
output_path = "s3://sagemaker-us-east-2-{}/{}".format(
DEV_ACCOUNT, "specified_ouput_path")
problem_type = "MulticlassClassification"
job_objective = {"MetricName": "Accuracy"}
auto_ml = AutoML(
role=ROLE,
target_attribute_name=TARGET_ATTRIBUTE_NAME,
sagemaker_session=sagemaker_session,
max_candidates=1,
output_path=output_path,
problem_type=problem_type,
job_objective=job_objective,
)
inputs = TRAINING_DATA
with timeout(minutes=AUTO_ML_DEFAULT_TIMEMOUT_MINUTES):
auto_ml.fit(inputs, job_name=JOB_NAME)
auto_ml_desc = auto_ml.describe_auto_ml_job(job_name=JOB_NAME)
assert auto_ml_desc["AutoMLJobStatus"] == "Completed"
assert auto_ml_desc["AutoMLJobName"] == JOB_NAME
assert auto_ml_desc["AutoMLJobObjective"] == job_objective
assert auto_ml_desc["ProblemType"] == problem_type
assert auto_ml_desc["OutputDataConfig"]["S3OutputPath"] == output_path
