Skip to content

Use the test argement framework_version in all tests #158

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jan 29, 2019
Merged

Use the test argement framework_version in all tests #158

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7cea4aa
Use the test argement framework_version in all tests
icywang86rui Jan 29, 2019
8e63a55
Make flake8 happy
icywang86rui Jan 29, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion test/integration/local/test_horovod.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,7 +32,8 @@ def test_distributed_training_horovod_basic(instances,
processes,
sagemaker_local_session,
docker_image,
tmpdir):
tmpdir,
framework_version):
output_path = 'file://%s' % tmpdir
estimator = TensorFlow(
entry_point=os.path.join(RESOURCE_PATH, 'hvdbasic', 'train_hvd_basic.py'),
Expand All @@ -42,6 +43,7 @@ def test_distributed_training_horovod_basic(instances,
train_instance_count=instances,
image_name=docker_image,
output_path=output_path,
framework_version=framework_version,
hyperparameters={'sagemaker_mpi_enabled': True,
'sagemaker_network_interface_name': 'eth0',
'sagemaker_mpi_num_of_processes_per_host': processes})
Expand Down
6 changes: 3 additions & 3 deletions test/integration/local/test_keras.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@


@pytest.mark.skip_gpu
def test_keras_training(sagemaker_local_session, docker_image, tmpdir):
def test_keras_training(sagemaker_local_session, docker_image, tmpdir, framework_version):
entry_point = os.path.join(RESOURCE_PATH, 'keras_inception.py')
output_path = 'file://{}'.format(tmpdir)

Expand All @@ -39,14 +39,14 @@ def test_keras_training(sagemaker_local_session, docker_image, tmpdir):
sagemaker_session=sagemaker_local_session,
model_dir='/opt/ml/model',
output_path=output_path,
framework_version='1.11.0',
framework_version=framework_version,
py_version='py3')

estimator.fit()

model = serving.Model(model_data=output_path,
role='SageMakerRole',
framework_version='1.11.0',
framework_version=framework_version,
sagemaker_session=sagemaker_local_session)

predictor = model.deploy(initial_instance_count=1, instance_type='local')
Expand Down
25 changes: 19 additions & 6 deletions test/integration/local/test_training.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,52 +39,62 @@ def test_py_versions(docker_image, processor, py_full_version):


@pytest.mark.skip_gpu
def test_mnist_cpu(sagemaker_local_session, docker_image, tmpdir):
def test_mnist_cpu(sagemaker_local_session, docker_image, tmpdir, framework_version):
output_path = 'file://{}'.format(tmpdir)
run_tf_training(script=os.path.join(RESOURCE_PATH, 'mnist', 'mnist.py'),
instance_type='local',
instance_count=1,
sagemaker_local_session=sagemaker_local_session,
docker_image=docker_image,
framework_version=framework_version,
output_path=output_path,
training_data_path='file://{}'.format(
os.path.join(RESOURCE_PATH, 'mnist', 'data')))
_assert_files_exist_in_tar(output_path, ['my_model.h5'])


@pytest.mark.skip_cpu
def test_gpu(sagemaker_local_session, docker_image):
def test_gpu(sagemaker_local_session, docker_image, framework_version):
run_tf_training(script=os.path.join(RESOURCE_PATH, 'gpu_device_placement.py'),
instance_type='local_gpu',
instance_count=1,
sagemaker_local_session=sagemaker_local_session,
docker_image=docker_image,
framework_version=framework_version,
training_data_path='file://{}'.format(
os.path.join(RESOURCE_PATH, 'mnist', 'data')))


@pytest.mark.skip_gpu
def test_distributed_training_cpu_no_ps(sagemaker_local_session, docker_image, tmpdir):
def test_distributed_training_cpu_no_ps(sagemaker_local_session,
docker_image,
tmpdir,
framework_version):
output_path = 'file://{}'.format(tmpdir)
run_tf_training(script=os.path.join(RESOURCE_PATH, 'mnist', 'mnist_estimator.py'),
instance_type='local',
instance_count=2,
sagemaker_local_session=sagemaker_local_session,
docker_image=docker_image,
framework_version=framework_version,
output_path=output_path,
training_data_path='file://{}'.format(
os.path.join(RESOURCE_PATH, 'mnist', 'data-distributed')))
_assert_files_exist_in_tar(output_path, TF_CHECKPOINT_FILES)


@pytest.mark.skip_gpu
def test_distributed_training_cpu_ps(sagemaker_local_session, docker_image, tmpdir):
def test_distributed_training_cpu_ps(sagemaker_local_session,
docker_image,
tmpdir,
framework_version):
output_path = 'file://{}'.format(tmpdir)
run_tf_training(script=os.path.join(RESOURCE_PATH, 'mnist', 'mnist_estimator.py'),
instance_type='local',
instance_count=2,
sagemaker_local_session=sagemaker_local_session,
docker_image=docker_image,
framework_version=framework_version,
output_path=output_path,
hyperparameters={'sagemaker_parameter_server_enabled': True},
training_data_path='file://{}'.format(
Expand All @@ -96,7 +106,10 @@ def run_tf_training(script,
instance_type,
instance_count,
sagemaker_local_session,
docker_image, training_data_path, output_path=None,
docker_image,
framework_version,
training_data_path,
output_path=None,
hyperparameters=None):

hyperparameters = hyperparameters or {}
Expand All @@ -111,7 +124,7 @@ def run_tf_training(script,
output_path=output_path,
hyperparameters=hyperparameters,
base_job_name='test-tf',
framework_version='1.11.0',
framework_version=framework_version,
py_version='py3')

estimator.fit(training_data_path)
Expand Down
5 changes: 3 additions & 2 deletions test/integration/sagemaker/test_horovod.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,7 +24,8 @@ def test_distributed_training_horovod(sagemaker_session,
sagemaker_local_session,
instance_type,
ecr_image,
tmpdir):
tmpdir,
framework_version):

mpi_options = '-verbose -x orte_base_help_aggregate=0'
estimator = TensorFlow(
Expand All @@ -33,7 +34,7 @@ def test_distributed_training_horovod(sagemaker_session,
train_instance_type=instance_type,
train_instance_count=2,
image_name=ecr_image,
framework_version='1.12',
framework_version=framework_version,
py_version='py3',
script_mode=True,
hyperparameters={'sagemaker_mpi_enabled': True,
Expand Down
18 changes: 9 additions & 9 deletions test/integration/sagemaker/test_mnist.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@
from sagemaker_tensorflow_container.training import SAGEMAKER_PARAMETER_SERVER_ENABLED


def test_mnist(sagemaker_session, ecr_image, instance_type):
def test_mnist(sagemaker_session, ecr_image, instance_type, framework_version):
resource_path = os.path.join(os.path.dirname(__file__), '../..', 'resources')
script = os.path.join(resource_path, 'mnist', 'mnist.py')
estimator = TensorFlow(entry_point=script,
Expand All @@ -30,7 +30,7 @@ def test_mnist(sagemaker_session, ecr_image, instance_type):
train_instance_count=1,
sagemaker_session=sagemaker_session,
image_name=ecr_image,
framework_version='1.12.0',
framework_version=framework_version,
py_version='py3',
base_job_name='test-sagemaker-mnist')
inputs = estimator.sagemaker_session.upload_data(
Expand All @@ -40,7 +40,7 @@ def test_mnist(sagemaker_session, ecr_image, instance_type):
_assert_s3_file_exists(estimator.model_data)


def test_distributed_mnist_no_ps(sagemaker_session, ecr_image, instance_type):
def test_distributed_mnist_no_ps(sagemaker_session, ecr_image, instance_type, framework_version):
resource_path = os.path.join(os.path.dirname(__file__), '../..', 'resources')
script = os.path.join(resource_path, 'mnist', 'mnist.py')
estimator = TensorFlow(entry_point=script,
Expand All @@ -49,7 +49,7 @@ def test_distributed_mnist_no_ps(sagemaker_session, ecr_image, instance_type):
train_instance_type=instance_type,
sagemaker_session=sagemaker_session,
image_name=ecr_image,
framework_version='1.12.0',
framework_version=framework_version,
py_version='py3',
base_job_name='test-tf-sm-distributed-mnist')
inputs = estimator.sagemaker_session.upload_data(
Expand All @@ -59,7 +59,7 @@ def test_distributed_mnist_no_ps(sagemaker_session, ecr_image, instance_type):
_assert_s3_file_exists(estimator.model_data)


def test_distributed_mnist_ps(sagemaker_session, ecr_image, instance_type):
def test_distributed_mnist_ps(sagemaker_session, ecr_image, instance_type, framework_version):
resource_path = os.path.join(os.path.dirname(__file__), '..', '..', 'resources')
script = os.path.join(resource_path, 'mnist', 'mnist_estimator.py')
estimator = TensorFlow(entry_point=script,
Expand All @@ -69,7 +69,7 @@ def test_distributed_mnist_ps(sagemaker_session, ecr_image, instance_type):
train_instance_type=instance_type,
sagemaker_session=sagemaker_session,
image_name=ecr_image,
framework_version='1.12.0',
framework_version=framework_version,
py_version='py3',
base_job_name='test-tf-sm-distributed-mnist')
inputs = estimator.sagemaker_session.upload_data(
Expand All @@ -80,7 +80,7 @@ def test_distributed_mnist_ps(sagemaker_session, ecr_image, instance_type):
_assert_s3_file_exists(estimator.model_data)


def test_s3_plugin(sagemaker_session, ecr_image, instance_type, region):
def test_s3_plugin(sagemaker_session, ecr_image, instance_type, region, framework_version):
resource_path = os.path.join(os.path.dirname(__file__), '..', '..', 'resources')
script = os.path.join(resource_path, 'mnist', 'mnist_estimator.py')
estimator = TensorFlow(entry_point=script,
Expand All @@ -91,7 +91,7 @@ def test_s3_plugin(sagemaker_session, ecr_image, instance_type, region):
# Disable throttling for checkpoint and model saving
'throttle-secs': 0,
# Without the patch training jobs would fail around 100th to
# 150th steps
# 150th step
'max-steps': 200,
# Large batch size would result in a larger checkpoint file
'batch-size': 2048,
Expand All @@ -103,7 +103,7 @@ def test_s3_plugin(sagemaker_session, ecr_image, instance_type, region):
train_instance_type=instance_type,
sagemaker_session=sagemaker_session,
image_name=ecr_image,
framework_version='1.12.0',
framework_version=framework_version,
py_version='py3',
base_job_name='test-tf-sm-s3-mnist')
estimator.fit('s3://sagemaker-sample-data-{}/tensorflow/mnist'.format(region))
Expand Down