breaking: require framework_version, py_version for mxnet

* framework_version, py_version required for framework MXNet
* framework_version, py_version required for framework MXNetModel
* re-order of non-default args, convention to follow entry_point
* testing updates
* doc updates
* ignore coverage results for py27 env due to v2 migration scripts

Author: metrizable

doc/frameworks/mxnet/using_mxnet.rst

@@ -345,7 +345,8 @@ The following code sample shows how you train a custom MXNet script "train.py".
     mxnet_estimator = MXNet('train.py',
                             train_instance_type='ml.p2.xlarge',
                             train_instance_count=1,
-                            framework_version='1.3.0',
+                            framework_version='1.6.0',
+                            py_version='py3',
                             hyperparameters={'batch-size': 100,
                                              'epochs': 10,
                                              'learning-rate': 0.1})
@@ -392,10 +393,10 @@ If you use the ``MXNet`` estimator to train the model, you can call ``deploy`` t
 
     # Train my estimator
     mxnet_estimator = MXNet('train.py',
-                            train_instance_type='ml.p2.xlarge',
-                            train_instance_count=1,
+                            framework_version='1.6.0',
                             py_version='py3',
-                            framework_version='1.6.0')
+                            train_instance_type='ml.p2.xlarge',
+                            train_instance_count=1)
     mxnet_estimator.fit('s3://my_bucket/my_training_data/')
 
     # Deploy my estimator to an Amazon SageMaker Endpoint and get a Predictor
@@ -409,8 +410,8 @@ If using a pretrained model, create an ``MXNetModel`` object, and then call ``de
     mxnet_model = MXNetModel(model_data='s3://my_bucket/pretrained_model/model.tar.gz',
                              role=role,
                              entry_point='inference.py',
-                             py_version='py3',
-                             framework_version='1.6.0')
+                             framework_version='1.6.0',
+                             py_version='py3')
     predictor = mxnet_model.deploy(instance_type='ml.m4.xlarge',
                                    initial_instance_count=1)
 

src/sagemaker/cli/mxnet.py

@@ -14,6 +14,7 @@
 from __future__ import absolute_import
 
 from sagemaker.cli.common import HostCommand, TrainCommand
+from sagemaker.mxnet import defaults
 
 
 def train(args):
@@ -40,13 +41,14 @@ def create_estimator(self):
         from sagemaker.mxnet.estimator import MXNet
 
         return MXNet(
-            self.script,
+            entry_point=self.script,
+            framework_version=defaults.MXNET_VERSION,
+            py_version=self.python,
             role=self.role_name,
             base_job_name=self.job_name,
             train_instance_count=self.instance_count,
             train_instance_type=self.instance_type,
             hyperparameters=self.hyperparameters,
-            py_version=self.python,
         )
 
 
@@ -64,6 +66,7 @@ def create_model(self, model_url):
             model_data=model_url,
             role=self.role_name,
             entry_point=self.script,
+            framework_version=defaults.MXNET_VERSION,
             py_version=self.python,
             name=self.endpoint_name,
             env=self.environment,

src/sagemaker/mxnet/estimator.py

@@ -19,7 +19,6 @@
 from sagemaker.fw_utils import (
     framework_name_from_image,
     framework_version_from_tag,
-    empty_framework_version_warning,
     python_deprecation_warning,
     parameter_v2_rename_warning,
     is_version_equal_or_higher,
@@ -43,10 +42,10 @@ class MXNet(Framework):
     def __init__(
         self,
         entry_point,
+        framework_version,
+        py_version,
         source_dir=None,
         hyperparameters=None,
-        py_version="py2",
-        framework_version=None,
         image_name=None,
         distributions=None,
         **kwargs
@@ -73,6 +72,11 @@ def __init__(
                 file which should be executed as the entry point to training.
                 If ``source_dir`` is specified, then ``entry_point``
                 must point to a file located at the root of ``source_dir``.
+            framework_version (str): MXNet version you want to use for executing
+                your model training code. List of supported versions
+                https://github.com/aws/sagemaker-python-sdk#mxnet-sagemaker-estimators.
+            py_version (str): Python version you want to use for executing your
+                model training code. One of 'py2' or 'py3'.
             source_dir (str): Path (absolute, relative or an S3 URI) to a directory
                 with any other training source code dependencies aside from the entry
                 point file (default: None). If ``source_dir`` is an S3 URI, it must
@@ -84,12 +88,6 @@ def __init__(
                 SageMaker. For convenience, this accepts other types for keys
                 and values, but ``str()`` will be called to convert them before
                 training.
-            py_version (str): Python version you want to use for executing your
-                model training code (default: 'py2'). One of 'py2' or 'py3'.
-            framework_version (str): MXNet version you want to use for executing
-                your model training code. List of supported versions
-                https://github.com/aws/sagemaker-python-sdk#mxnet-sagemaker-estimators.
-                If not specified, this will default to 1.2.1.
             image_name (str): If specified, the estimator will use this image for training and
                 hosting, instead of selecting the appropriate SageMaker official image based on
                 framework_version and py_version. It can be an ECR url or dockerhub image and tag.
@@ -110,34 +108,31 @@ def __init__(
             :class:`~sagemaker.estimator.Framework` and
             :class:`~sagemaker.estimator.EstimatorBase`.
         """
-        if framework_version is None:
+        self.framework_version = framework_version
+        self.py_version = py_version
+        if self.py_version == "py2":
             logger.warning(
-                empty_framework_version_warning(defaults.MXNET_VERSION, self.LATEST_VERSION)
+                python_deprecation_warning(self.__framework_name__, defaults.LATEST_PY2_VERSION)
             )
-        self.framework_version = framework_version or defaults.MXNET_VERSION
 
         if "enable_sagemaker_metrics" not in kwargs:
             # enable sagemaker metrics for MXNet v1.6 or greater:
-            if is_version_equal_or_higher([1, 6], self.framework_version):
+            if self.framework_version and is_version_equal_or_higher(
+                [1, 6], self.framework_version
+            ):
                 kwargs["enable_sagemaker_metrics"] = True
 
         super(MXNet, self).__init__(
             entry_point, source_dir, hyperparameters, image_name=image_name, **kwargs
         )
 
-        if py_version == "py2":
-            logger.warning(
-                python_deprecation_warning(self.__framework_name__, defaults.LATEST_PY2_VERSION)
-            )
-
         if distributions is not None:
             logger.warning(parameter_v2_rename_warning("distributions", "distribution"))
             train_instance_type = kwargs.get("train_instance_type")
             warn_if_parameter_server_with_multi_gpu(
                 training_instance_type=train_instance_type, distributions=distributions
             )
 
-        self.py_version = py_version
         self._configure_distribution(distributions)
 
     def _configure_distribution(self, distributions):
@@ -221,12 +216,12 @@ def create_model(
             self.model_data,
             role or self.role,
             entry_point or self.entry_point,
+            framework_version=self.framework_version,
+            py_version=self.py_version,
             source_dir=(source_dir or self._model_source_dir()),
             enable_cloudwatch_metrics=self.enable_cloudwatch_metrics,
             container_log_level=self.container_log_level,
             code_location=self.code_location,
-            py_version=self.py_version,
-            framework_version=self.framework_version,
             model_server_workers=model_server_workers,
             sagemaker_session=self.sagemaker_session,
             vpc_config=self.get_vpc_config(vpc_config_override),
@@ -257,6 +252,8 @@ class constructor
         if not framework:
             # If we were unable to parse the framework name from the image it is not one of our
             # officially supported images, in this case just add the image to the init params.
+            init_params["framework_version"] = None
+            init_params["py_version"] = None
             init_params["image_name"] = image_name
             return init_params
 

src/sagemaker/mxnet/model.py

@@ -18,12 +18,7 @@
 import packaging.version
 
 import sagemaker
-from sagemaker.fw_utils import (
-    create_image_uri,
-    model_code_key_prefix,
-    python_deprecation_warning,
-    empty_framework_version_warning,
-)
+from sagemaker.fw_utils import create_image_uri, model_code_key_prefix, python_deprecation_warning
 from sagemaker.model import FrameworkModel, MODEL_SERVER_WORKERS_PARAM_NAME
 from sagemaker.mxnet import defaults
 from sagemaker.predictor import RealTimePredictor, json_serializer, json_deserializer
@@ -65,9 +60,9 @@ def __init__(
         model_data,
         role,
         entry_point,
+        framework_version,
+        py_version,
         image=None,
-        py_version="py2",
-        framework_version=None,
         predictor_cls=MXNetPredictor,
         model_server_workers=None,
         **kwargs
@@ -86,12 +81,12 @@ def __init__(
                 file which should be executed as the entry point to model
                 hosting. If ``source_dir`` is specified, then ``entry_point``
                 must point to a file located at the root of ``source_dir``.
-            image (str): A Docker image URI (default: None). If not specified, a
-                default image for MXNet will be used.
-            py_version (str): Python version you want to use for executing your
-                model training code (default: 'py2').
             framework_version (str): MXNet version you want to use for executing
                 your model training code.
+            py_version (str): Python version you want to use for executing your
+                model training code.
+            image (str): A Docker image URI (default: None). If not specified, a
+                default image for MXNet will be used.
             predictor_cls (callable[str, sagemaker.session.Session]): A function
                 to call to create a predictor with an endpoint name and
                 SageMaker ``Session``. If specified, ``deploy()`` returns the
@@ -108,22 +103,17 @@ def __init__(
             :class:`~sagemaker.model.FrameworkModel` and
             :class:`~sagemaker.model.Model`.
         """
-        super(MXNetModel, self).__init__(
-            model_data, image, role, entry_point, predictor_cls=predictor_cls, **kwargs
-        )
-
-        if py_version == "py2":
+        self.framework_version = framework_version
+        self.py_version = py_version
+        if self.py_version == "py2":
             logger.warning(
                 python_deprecation_warning(self.__framework_name__, defaults.LATEST_PY2_VERSION)
             )
 
-        if framework_version is None:
-            logger.warning(
-                empty_framework_version_warning(defaults.MXNET_VERSION, defaults.LATEST_VERSION)
-            )
+        super(MXNetModel, self).__init__(
+            model_data, image, role, entry_point, predictor_cls=predictor_cls, **kwargs
+        )
 
-        self.py_version = py_version
-        self.framework_version = framework_version or defaults.MXNET_VERSION
         self.model_server_workers = model_server_workers
 
     def prepare_container_def(self, instance_type, accelerator_type=None):

tests/conftest.py

@@ -163,6 +163,11 @@ def mxnet_version(request):
     return request.param
 
 
+@pytest.fixture(scope="module", params=["py2", "py3"])
+def mxnet_py_version(request):
+    return request.param
+
+
 @pytest.fixture(scope="module", params=["0.4", "0.4.0", "1.0", "1.0.0"])
 def pytorch_version(request):
     return request.param

tests/unit/test_multidatamodel.py

@@ -27,7 +27,11 @@
 MXNET_MODEL_DATA = "s3://mybucket/mxnet_path/model.tar.gz"
 MXNET_MODEL_NAME = "dummy-mxnet-model"
 MXNET_ROLE = "DummyMXNetRole"
-MXNET_IMAGE = "520713654638.dkr.ecr.us-west-2.amazonaws.com/sagemaker-mxnet:1.2-cpu-py2"
+MXNET_FRAMEWORK_VERSION = "1.2"
+MXNET_PY_VERSION = "py2"
+MXNET_IMAGE = "520713654638.dkr.ecr.us-west-2.amazonaws.com/sagemaker-mxnet:{}-cpu-{}".format(
+    MXNET_FRAMEWORK_VERSION, MXNET_PY_VERSION
+)
 
 DATA_DIR = os.path.join(os.path.dirname(__file__), "..", "data")
 IMAGE = "123456789012.dkr.ecr.dummyregion.amazonaws.com/dummyimage:latest"
@@ -100,8 +104,10 @@ def multi_data_model(sagemaker_session):
 def mxnet_model(sagemaker_session):
     return MXNetModel(
         MXNET_MODEL_DATA,
-        role=MXNET_ROLE,
         entry_point=ENTRY_POINT,
+        framework_version=MXNET_FRAMEWORK_VERSION,
+        py_version=MXNET_PY_VERSION,
+        role=MXNET_ROLE,
         sagemaker_session=sagemaker_session,
         name=MXNET_MODEL_NAME,
         enable_network_isolation=True,