Merge pull request aws#9 from verdimrc/fp-revert-to-code

FrameworkProcessor: revert `run(entry_point=...)` back to `run(code=...)`

Author: verdimrc

src/sagemaker/mxnet/processing.py

@@ -50,7 +50,7 @@ def __init__(
 
         Unless ``image_uri`` is specified, the MXNet environment is an
         Amazon-built Docker container that executes functions defined in the supplied
-        ``entry_point`` Python script.
+        ``code`` Python script.
 
         The arguments have the exact same meaning as in ``FrameworkProcessor``.
 

src/sagemaker/processing.py

@@ -1264,7 +1264,7 @@ def __init__(
             image_uri (str): The URI of the Docker image to use for the
                 processing jobs (default: None).
             command ([str]): The command to run, along with any command-line flags
-                to *precede* the ```entry_point script``` (default: ['python']).
+                to *precede* the ```code script``` (default: ['python']).
             volume_size_in_gb (int): Size in GB of the EBS volume
                 to use for storing data during processing (default: 30).
             volume_kms_key (str): A KMS key for the processing volume (default: None).
@@ -1361,7 +1361,7 @@ def _pre_init_normalization(
 
     def run(  # type: ignore[override]
         self,
-        entry_point,
+        code,
         source_dir=None,
         dependencies=None,
         git_config=None,
@@ -1377,9 +1377,9 @@ def run(  # type: ignore[override]
         """Runs a processing job.
 
         Args:
-            entrypoint (str): Path (absolute or relative) to the local Python source
+            code (str): Path (absolute or relative) to the local Python source
                 file which should be executed as the entry point to training. If
-                ``source_dir`` is specified, then ``entry_point`` must point to a file
+                ``source_dir`` is specified, then ``code`` must point to a file
                 located at the root of ``source_dir``.
             source_dir (str): Path (absolute, relative or an S3 URI) to a directory
                 with any other training source code dependencies aside from the entry
@@ -1464,9 +1464,7 @@ def run(  # type: ignore[override]
         if job_name is None:
             job_name = self._generate_current_job_name()
 
-        estimator = self._upload_payload(
-            entry_point, source_dir, dependencies, git_config, job_name
-        )
+        estimator = self._upload_payload(code, source_dir, dependencies, git_config, job_name)
         inputs = self._patch_inputs_with_payload(
             inputs, estimator._hyperparameters["sagemaker_submit_directory"]
         )
@@ -1511,13 +1509,14 @@ def _generate_framework_script(self, user_script: str) -> str:
         This script implements the "framework" functionality for setting up your code:
         Untar-ing the sourcedir bundle in the ```code``` input; installing extra
         runtime dependencies if specified; and then invoking the ```command``` and
-        ```entry_point``` configured for the job.
+        ```code``` configured for the job.
 
         Args:
-            user_script (str): Relative path to ```entry_point``` in the source bundle
+            user_script (str): Relative path to ```code``` in the source bundle
                 - e.g. 'process.py'.
         """
-        return dedent("""\
+        return dedent(
+            """\
             #!/bin/bash
 
             cd /opt/ml/processing/input/code/
@@ -1529,7 +1528,8 @@ def _generate_framework_script(self, user_script: str) -> str:
             [[ -f 'requirements.txt' ]] && pip install -r requirements.txt
 
             {entry_point_command} {entry_point} "$@"
-        """).format(
+        """
+        ).format(
             entry_point_command=" ".join(self.command),
             entry_point=user_script,
         )
@@ -1605,7 +1605,7 @@ def _set_entrypoint(self, command, user_script_name):
             command ([str]): Ignored in favor of self.framework_entrypoint_command
             user_script_name (str): A filename with an extension.
         """
-        
+
         user_script_location = str(
             pathlib.PurePosixPath(
                 self._CODE_CONTAINER_BASE_PATH, self._CODE_CONTAINER_INPUT_NAME, user_script_name

src/sagemaker/pytorch/processing.py

@@ -50,7 +50,7 @@ def __init__(
 
         Unless ``image_uri`` is specified, the PyTorch environment is an
         Amazon-built Docker container that executes functions defined in the supplied
-        ``entry_point`` Python script.
+        ``code`` Python script.
 
         The arguments have the exact same meaning as in ``FrameworkProcessor``.
 

src/sagemaker/sklearn/processing.py

@@ -28,7 +28,7 @@ class SKLearnProcessor(FrameworkProcessor):
 
     Unless ``image_uri`` is specified, the scikit-learn environment is an
     Amazon-built Docker container that executes functions defined in the supplied
-    ``entry_point`` Python script.
+    ``code`` Python script.
 
     The arguments have the exact same meaning as in ``FrameworkProcessor``.
 

src/sagemaker/tensorflow/processing.py

@@ -50,7 +50,7 @@ def __init__(
 
         Unless ``image_uri`` is specified, the TensorFlow environment is an
         Amazon-built Docker container that executes functions defined in the supplied
-        ``entry_point`` Python script.
+        ``code`` Python script.
 
         The arguments have the exact same meaning as in ``FrameworkProcessor``.
 

src/sagemaker/xgboost/processing.py

@@ -49,7 +49,7 @@ def __init__(
         """This processor executes a Python script in an XGBoost execution environment.
 
         Unless ``image_uri`` is specified, the XGBoost environment is an Amazon-built
-        Docker container that executes functions defined in the supplied ``entry_point``
+        Docker container that executes functions defined in the supplied ``code``
         Python script.
 
         The arguments have the exact same meaning as in ``FrameworkProcessor``.

tests/integ/test_processing.py

@@ -130,7 +130,7 @@ def test_sklearn(sagemaker_session, sklearn_latest_version, cpu_instance_type):
     )
 
     sklearn_processor.run(
-        entry_point=script_path,
+        code=script_path,
         inputs=[ProcessingInput(source=input_file_path, destination="/opt/ml/processing/inputs/")],
         wait=False,
         logs=False,
@@ -174,7 +174,7 @@ def test_sklearn_with_customizations(
     )
 
     sklearn_processor.run(
-        entry_point=os.path.join(DATA_DIR, "dummy_script.py"),
+        code=os.path.join(DATA_DIR, "dummy_script.py"),
         inputs=[
             ProcessingInput(
                 source=input_file_path,
@@ -258,7 +258,7 @@ def test_sklearn_with_custom_default_bucket(
     )
 
     sklearn_processor.run(
-        entry_point=os.path.join(DATA_DIR, "dummy_script.py"),
+        code=os.path.join(DATA_DIR, "dummy_script.py"),
         inputs=[
             ProcessingInput(
                 source=input_file_path,
@@ -338,7 +338,7 @@ def test_sklearn_with_no_inputs_or_outputs(
     )
 
     sklearn_processor.run(
-        entry_point=os.path.join(DATA_DIR, "dummy_script.py"),
+        code=os.path.join(DATA_DIR, "dummy_script.py"),
         arguments=["-v"],
         wait=True,
         logs=True,
@@ -695,7 +695,7 @@ def test_sklearn_with_network_config(sagemaker_session, sklearn_latest_version,
     )
 
     sklearn_processor.run(
-        entry_point=script_path,
+        code=script_path,
         inputs=[ProcessingInput(source=input_file_path, destination="/opt/ml/processing/inputs/")],
         wait=False,
         logs=False,

tests/unit/test_processing.py

@@ -92,7 +92,7 @@ def test_sklearn_processor_with_required_parameters(
         sagemaker_session=sagemaker_session,
     )
 
-    processor.run(entry_point="/local/path/to/processing_code.py")
+    processor.run(code="/local/path/to/processing_code.py")
 
     expected_args = _get_expected_args_modular_code(processor._current_job_name)
 
@@ -137,7 +137,7 @@ def test_sklearn_with_all_parameters(
 
     with patch("sagemaker.estimator.tar_and_upload_dir", return_value=uploaded_code):
         processor.run(
-            entry_point="processing_code.py",
+            code="processing_code.py",
             source_dir="/local/path/to/source_dir",
             dependencies=["/local/path/to/dep_01"],
             inputs=_get_data_inputs_all_parameters(),
@@ -200,7 +200,7 @@ def test_sklearn_with_all_parameters_via_run_args(
     )
 
     processor.run(
-        entry_point=run_args.code,
+        code=run_args.code,
         inputs=run_args.inputs,
         outputs=run_args.outputs,
         arguments=run_args.arguments,
@@ -223,11 +223,11 @@ def test_sklearn_with_all_parameters_via_run_args(
     # Verify the alternate command was applied successfully:
     framework_script = processor._generate_framework_script("processing_code.py")
     expected_invocation = f"{' '.join(custom_command)} processing_code.py"
-    assert f"\n{expected_invocation}" in framework_script, (
-        "Framework script should contain customized invocation:\n{}\n\nGot:\n{}".format(
-            expected_invocation,
-            framework_script,
-        )
+    assert (
+        f"\n{expected_invocation}" in framework_script
+    ), "Framework script should contain customized invocation:\n{}\n\nGot:\n{}".format(
+        expected_invocation,
+        framework_script,
     )
 
 
@@ -276,7 +276,7 @@ def test_sklearn_with_all_parameters_via_run_args_called_twice(
     )
 
     processor.run(
-        entry_point=run_args.code,
+        code=run_args.code,
         inputs=run_args.inputs,
         outputs=run_args.outputs,
         arguments=run_args.arguments,
@@ -838,7 +838,9 @@ def _get_data_outputs_all_parameters():
     ]
 
 
-def _get_expected_args_all_parameters_modular_code(job_name, code_s3_uri=MOCKED_S3_URI, instance_count=1):
+def _get_expected_args_all_parameters_modular_code(
+    job_name, code_s3_uri=MOCKED_S3_URI, instance_count=1
+):
     # Add something to inputs
     return {
         "inputs": [