def __rcf_training_job(
sagemaker_session, container_image, cpu_instance_type, num_trees, num_samples_per_tree
):
job_name = unique_name_from_base("randomcutforest")
with timeout(minutes=TRAINING_DEFAULT_TIMEOUT_MINUTES):
# Generate a thousand 14-dimensional datapoints.
feature_num = 14
train_input = numpy.random.rand(1000, feature_num)
rcf = RandomCutForest(
role=ROLE,
instance_count=1,
instance_type=cpu_instance_type,
num_trees=num_trees,
num_samples_per_tree=num_samples_per_tree,
eval_metrics=["accuracy", "precision_recall_fscore"],
sagemaker_session=sagemaker_session,
)
rcf.fit(records=rcf.record_set(train_input), job_name=job_name)
# Replace the container image value with a multi-model container image for now since the
# frameworks do not support multi-model container image yet.
rcf_model = rcf.create_model()
rcf_model.image_uri = container_image
return rcf_model
def test_model_image(sagemaker_session):
randomcutforest = RandomCutForest(sagemaker_session=sagemaker_session, **ALL_REQ_ARGS)
data = RecordSet("s3://{}/{}".format(BUCKET_NAME, PREFIX), num_records=1, feature_dim=FEATURE_DIM, channel='train')
randomcutforest.fit(data, MINI_BATCH_SIZE)
model = randomcutforest.create_model()
assert model.image == registry(REGION, "randomcutforest") + '/randomcutforest:1'
def test_model_image(sagemaker_session):
randomcutforest = RandomCutForest(sagemaker_session=sagemaker_session, **ALL_REQ_ARGS)
data = RecordSet("s3://{}/{}".format(BUCKET_NAME, PREFIX), num_records=1, feature_dim=FEATURE_DIM, channel='train')
randomcutforest.fit(data, MINI_BATCH_SIZE)
model = randomcutforest.create_model()
assert model.image == registry(REGION, "randomcutforest") + '/randomcutforest:1'
def test_predictor_type(sagemaker_session):
randomcutforest = RandomCutForest(sagemaker_session=sagemaker_session, **ALL_REQ_ARGS)
data = RecordSet("s3://{}/{}".format(BUCKET_NAME, PREFIX), num_records=1, feature_dim=FEATURE_DIM, channel='train')
randomcutforest.fit(data, MINI_BATCH_SIZE)
model = randomcutforest.create_model()
predictor = model.deploy(1, TRAIN_INSTANCE_TYPE)
assert isinstance(predictor, RandomCutForestPredictor)
def test_predictor_type(sagemaker_session):
randomcutforest = RandomCutForest(sagemaker_session=sagemaker_session, **ALL_REQ_ARGS)
data = RecordSet("s3://{}/{}".format(BUCKET_NAME, PREFIX), num_records=1, feature_dim=FEATURE_DIM, channel='train')
randomcutforest.fit(data, MINI_BATCH_SIZE)
model = randomcutforest.create_model()
predictor = model.deploy(1, TRAIN_INSTANCE_TYPE)
assert isinstance(predictor, RandomCutForestPredictor)
def test_call_fit_none_mini_batch_size(sagemaker_session):
randomcutforest = RandomCutForest(base_job_name="randomcutforest",
sagemaker_session=sagemaker_session,
**ALL_REQ_ARGS)
data = RecordSet("s3://{}/{}".format(BUCKET_NAME, PREFIX),
num_records=1,
feature_dim=FEATURE_DIM,
channel='train')
randomcutforest.fit(data)
def test_model_image(sagemaker_session):
randomcutforest = RandomCutForest(sagemaker_session=sagemaker_session, **ALL_REQ_ARGS)
data = RecordSet(
"s3://{}/{}".format(BUCKET_NAME, PREFIX),
num_records=1,
feature_dim=FEATURE_DIM,
channel="train",
)
randomcutforest.fit(data, MINI_BATCH_SIZE)
model = randomcutforest.create_model()
assert image_uris.retrieve("randomcutforest", REGION) == model.image_uri
def test_call_fit_feature_dim_greater_than_max_allowed(sagemaker_session):
randomcutforest = RandomCutForest(base_job_name="randomcutforest",
sagemaker_session=sagemaker_session,
**ALL_REQ_ARGS)
data = RecordSet("s3://{}/{}".format(BUCKET_NAME, PREFIX),
num_records=1,
feature_dim=MAX_FEATURE_DIM + 1,
channel='train')
with pytest.raises((TypeError, ValueError)):
randomcutforest.fit(data)
def test_call_fit_wrong_type_mini_batch_size(sagemaker_session):
randomcutforest = RandomCutForest(base_job_name="randomcutforest",
sagemaker_session=sagemaker_session,
**ALL_REQ_ARGS)
data = RecordSet("s3://{}/{}".format(BUCKET_NAME, PREFIX),
num_records=1,
feature_dim=FEATURE_DIM,
channel='train')
with pytest.raises((TypeError, ValueError)):
randomcutforest.fit(data, 1234)
def test_call_fit(base_fit, sagemaker_session):
randomcutforest = RandomCutForest(base_job_name="randomcutforest", sagemaker_session=sagemaker_session,
**ALL_REQ_ARGS)
data = RecordSet("s3://{}/{}".format(BUCKET_NAME, PREFIX), num_records=1, feature_dim=FEATURE_DIM, channel='train')
randomcutforest.fit(data, MINI_BATCH_SIZE)
base_fit.assert_called_once()
assert len(base_fit.call_args[0]) == 2
assert base_fit.call_args[0][0] == data
assert base_fit.call_args[0][1] == MINI_BATCH_SIZE
def test_call_fit(base_fit, sagemaker_session):
randomcutforest = RandomCutForest(base_job_name="randomcutforest", sagemaker_session=sagemaker_session,
**ALL_REQ_ARGS)
data = RecordSet("s3://{}/{}".format(BUCKET_NAME, PREFIX), num_records=1, feature_dim=FEATURE_DIM, channel='train')
randomcutforest.fit(data, MINI_BATCH_SIZE)
base_fit.assert_called_once()
assert len(base_fit.call_args[0]) == 2
assert base_fit.call_args[0][0] == data
assert base_fit.call_args[0][1] == MINI_BATCH_SIZE
