applied style formatting to fix build issues

Author: aws-byeldos

src/sagemaker/clarify.py

@@ -924,6 +924,7 @@ def __init__(
             version (str): Clarify version to use.
         """  # noqa E501  # pylint: disable=c0301
         container_uri = image_uris.retrieve("clarify", sagemaker_session.boto_region_name, version)
+        self._last_analysis_config = None
         self.job_name_prefix = job_name_prefix
         super(SageMakerClarifyProcessor, self).__init__(
             role,
@@ -987,7 +988,7 @@ def _run(
         """
         # for debugging: to access locally, i.e. without a need to look for it in an S3 bucket
         self._last_analysis_config = analysis_config
-        logger.info("Analysis Config: ", analysis_config)
+        logger.info("Analysis Config: %s", analysis_config)
 
         with tempfile.TemporaryDirectory() as tmpdirname:
             analysis_config_file = os.path.join(tmpdirname, "analysis_config.json")
@@ -1086,12 +1087,12 @@ def run_pre_training_bias(
                 * ``'TrialComponentDisplayName'`` is used for display in Amazon SageMaker Studio.
         """  # noqa E501  # pylint: disable=c0301
         analysis_config = _AnalysisConfigGenerator.bias_pre_training(
-            data_config,
-            data_bias_config,
-            methods
+            data_config, data_bias_config, methods
         )
         # when name is either not provided (is None) or an empty string ("")
-        job_name = job_name or utils.name_from_base(self.job_name_prefix or "Clarify-Pretraining-Bias")
+        job_name = job_name or utils.name_from_base(
+            self.job_name_prefix or "Clarify-Pretraining-Bias"
+        )
         return self._run(
             data_config,
             analysis_config,
@@ -1167,14 +1168,12 @@ def run_post_training_bias(
                 * ``'TrialComponentDisplayName'`` is used for display in Amazon SageMaker Studio.
         """  # noqa E501  # pylint: disable=c0301
         analysis_config = _AnalysisConfigGenerator.bias_post_training(
-            data_config,
-            data_bias_config,
-            model_predicted_label_config,
-            methods,
-            model_config
+            data_config, data_bias_config, model_predicted_label_config, methods, model_config
         )
         # when name is either not provided (is None) or an empty string ("")
-        job_name = job_name or utils.name_from_base(self.job_name_prefix or "Clarify-Posttraining-Bias")
+        job_name = job_name or utils.name_from_base(
+            self.job_name_prefix or "Clarify-Posttraining-Bias"
+        )
         return self._run(
             data_config,
             analysis_config,
@@ -1354,13 +1353,12 @@ def run_explainability(
                 * ``'TrialComponentDisplayName'`` is used for display in Amazon SageMaker Studio.
         """  # noqa E501  # pylint: disable=c0301
         analysis_config = _AnalysisConfigGenerator.explainability(
-            data_config,
-            model_config,
-            model_scores,
-            explainability_config
+            data_config, model_config, model_scores, explainability_config
         )
         # when name is either not provided (is None) or an empty string ("")
-        job_name = job_name or utils.name_from_base(self.job_name_prefix or "Clarify-Explainability")
+        job_name = job_name or utils.name_from_base(
+            self.job_name_prefix or "Clarify-Explainability"
+        )
         return self._run(
             data_config,
             analysis_config,
@@ -1376,6 +1374,7 @@ class _AnalysisConfigGenerator:
     """
     Creates analysis_config objects for different type of runs.
     """
+
     @classmethod
     def explainability(
         cls,
@@ -1384,6 +1383,7 @@ def explainability(
         model_scores: ModelPredictedLabelConfig,
         explainability_config: ExplainabilityConfig,
     ):
+        """Generates a config for Explainability"""
         analysis_config = data_config.get_config()
         predictor_config = model_config.get_predictor_config()
         if isinstance(model_scores, ModelPredictedLabelConfig):
@@ -1423,11 +1423,14 @@ def explainability(
         return cls._common(analysis_config)
 
     @classmethod
-    def bias_pre_training(cls, data_config: DataConfig, bias_config: BiasConfig, methods: Union[str, List[str]]):
+    def bias_pre_training(
+        cls, data_config: DataConfig, bias_config: BiasConfig, methods: Union[str, List[str]]
+    ):
+        """Generates a config for Bias Pre Training"""
         analysis_config = {
             **data_config.get_config(),
             **bias_config.get_config(),
-            "methods": {"pre_training_bias": {"methods": methods}}
+            "methods": {"pre_training_bias": {"methods": methods}},
         }
         return cls._common(analysis_config)
 
@@ -1440,14 +1443,18 @@ def bias_post_training(
         methods: Union[str, List[str]],
         model_config: ModelConfig,
     ):
+        """Generates a config for Bias Post Training"""
         analysis_config = {
             **data_config.get_config(),
             **bias_config.get_config(),
             "predictor": {**model_config.get_predictor_config()},
             "methods": {"post_training_bias": {"methods": methods}},
         }
         if model_predicted_label_config:
-            probability_threshold, predictor_config = model_predicted_label_config.get_predictor_config()
+            (
+                probability_threshold,
+                predictor_config,
+            ) = model_predicted_label_config.get_predictor_config()
             if predictor_config:
                 analysis_config["predictor"].update(predictor_config)
             _set(probability_threshold, "probability_threshold", analysis_config)
@@ -1463,24 +1470,29 @@ def bias(
         pre_training_methods: Union[str, List[str]] = "all",
         post_training_methods: Union[str, List[str]] = "all",
     ):
+        """Generates a config for Bias"""
         analysis_config = {
             **data_config.get_config(),
             **bias_config.get_config(),
             "predictor": model_config.get_predictor_config(),
             "methods": {
                 "pre_training_bias": {"methods": pre_training_methods},
                 "post_training_bias": {"methods": post_training_methods},
-            }
+            },
         }
         if model_predicted_label_config:
-            probability_threshold, predictor_config = model_predicted_label_config.get_predictor_config()
+            (
+                probability_threshold,
+                predictor_config,
+            ) = model_predicted_label_config.get_predictor_config()
             if predictor_config:
                 analysis_config["predictor"].update(predictor_config)
             _set(probability_threshold, "probability_threshold", analysis_config)
         return cls._common(analysis_config)
 
     @staticmethod
     def _common(analysis_config):
+        """Extends analysis config with common values"""
         analysis_config["methods"]["report"] = {
             "name": "report",
             "title": "Analysis Report",

tests/unit/test_clarify.py

@@ -766,8 +766,8 @@ def test_pre_training_bias(
             "facet": [{"name_or_index": "F1"}],
             "group_variable": "F2",
             "methods": {
-                'report': {'name': 'report', 'title': 'Analysis Report'},
-                "pre_training_bias": {"methods": "all"}
+                "report": {"name": "report", "title": "Analysis Report"},
+                "pre_training_bias": {"methods": "all"},
             },
         }
         mock_method.assert_called_with(
@@ -832,8 +832,8 @@ def test_post_training_bias(
             "facet": [{"name_or_index": "F1"}],
             "group_variable": "F2",
             "methods": {
-                'report': {'name': 'report', 'title': 'Analysis Report'},
-                "post_training_bias": {"methods": "all"}
+                "report": {"name": "report", "title": "Analysis Report"},
+                "post_training_bias": {"methods": "all"},
             },
             "predictor": {
                 "model_name": "xgboost-model",
@@ -993,7 +993,7 @@ def _run_test_explain(
                 "top_k_features": 10,
             }
         expected_analysis_config["methods"] = {
-            'report': {'name': 'report', 'title': 'Analysis Report'},
+            "report": {"name": "report", "title": "Analysis Report"},
             **expected_explanation_configs,
         }
         mock_method.assert_called_with(
@@ -1300,43 +1300,49 @@ def test_analysis_config_generator_for_explainability(data_config, model_config)
         model_scores,
         SHAPConfig(),
     )
-    expected = {'dataset_type': 'text/csv',
-                'headers': ['Label', 'F1', 'F2', 'F3', 'F4'],
-                'joinsource_name_or_index': 'F4',
-                'label': 'Label',
-                'methods': {
-                    'report': {'name': 'report', 'title': 'Analysis Report'},
-                    'shap': {'save_local_shap_values': True, 'use_logit': False}
-                },
-                'predictor': {'initial_instance_count': 1,
-                              'instance_type': 'ml.c5.xlarge',
-                              'label_headers': ['success'],
-                              'model_name': 'xgboost-model',
-                              'probability': 'pr'}}
+    expected = {
+        "dataset_type": "text/csv",
+        "headers": ["Label", "F1", "F2", "F3", "F4"],
+        "joinsource_name_or_index": "F4",
+        "label": "Label",
+        "methods": {
+            "report": {"name": "report", "title": "Analysis Report"},
+            "shap": {"save_local_shap_values": True, "use_logit": False},
+        },
+        "predictor": {
+            "initial_instance_count": 1,
+            "instance_type": "ml.c5.xlarge",
+            "label_headers": ["success"],
+            "model_name": "xgboost-model",
+            "probability": "pr",
+        },
+    }
     assert actual == expected
 
 
 def test_analysis_config_generator_for_bias_pre_training(data_config, data_bias_config):
     actual = _AnalysisConfigGenerator.bias_pre_training(
-        data_config,
-        data_bias_config,
-        methods="all"
+        data_config, data_bias_config, methods="all"
     )
-    expected = {'dataset_type': 'text/csv',
-                'facet': [{'name_or_index': 'F1'}],
-                'group_variable': 'F2',
-                'headers': ['Label', 'F1', 'F2', 'F3', 'F4'],
-                'joinsource_name_or_index': 'F4',
-                'label': 'Label',
-                'label_values_or_threshold': [1],
-                'methods': {
-                    'report': {'name': 'report', 'title': 'Analysis Report'},
-                    'pre_training_bias': {'methods': 'all'}}
-                }
+    expected = {
+        "dataset_type": "text/csv",
+        "facet": [{"name_or_index": "F1"}],
+        "group_variable": "F2",
+        "headers": ["Label", "F1", "F2", "F3", "F4"],
+        "joinsource_name_or_index": "F4",
+        "label": "Label",
+        "label_values_or_threshold": [1],
+        "methods": {
+            "report": {"name": "report", "title": "Analysis Report"},
+            "pre_training_bias": {"methods": "all"},
+        },
+    }
     assert actual == expected
 
 
-def test_analysis_config_generator_for_bias_post_training(data_config, data_bias_config, model_config):
+def test_analysis_config_generator_for_bias_post_training(
+    data_config, data_bias_config, model_config
+):
     model_predicted_label_config = ModelPredictedLabelConfig(
         probability="pr",
         label_headers=["success"],
@@ -1348,22 +1354,26 @@ def test_analysis_config_generator_for_bias_post_training(data_config, data_bias
         methods="all",
         model_config=model_config,
     )
-    expected = {'dataset_type': 'text/csv',
-                'facet': [{'name_or_index': 'F1'}],
-                'group_variable': 'F2',
-                'headers': ['Label', 'F1', 'F2', 'F3', 'F4'],
-                'joinsource_name_or_index': 'F4',
-                'label': 'Label',
-                'label_values_or_threshold': [1],
-                'methods': {
-                    'report': {'name': 'report', 'title': 'Analysis Report'},
-                    'post_training_bias': {'methods': 'all'}
-                },
-                'predictor': {'initial_instance_count': 1,
-                              'instance_type': 'ml.c5.xlarge',
-                              'label_headers': ['success'],
-                              'model_name': 'xgboost-model',
-                              'probability': 'pr'}}
+    expected = {
+        "dataset_type": "text/csv",
+        "facet": [{"name_or_index": "F1"}],
+        "group_variable": "F2",
+        "headers": ["Label", "F1", "F2", "F3", "F4"],
+        "joinsource_name_or_index": "F4",
+        "label": "Label",
+        "label_values_or_threshold": [1],
+        "methods": {
+            "report": {"name": "report", "title": "Analysis Report"},
+            "post_training_bias": {"methods": "all"},
+        },
+        "predictor": {
+            "initial_instance_count": 1,
+            "instance_type": "ml.c5.xlarge",
+            "label_headers": ["success"],
+            "model_name": "xgboost-model",
+            "probability": "pr",
+        },
+    }
     assert actual == expected
 
 
@@ -1380,20 +1390,25 @@ def test_analysis_config_generator_for_bias(data_config, data_bias_config, model
         pre_training_methods="all",
         post_training_methods="all",
     )
-    expected = {'dataset_type': 'text/csv',
-                'facet': [{'name_or_index': 'F1'}],
-                'group_variable': 'F2',
-                'headers': ['Label', 'F1', 'F2', 'F3', 'F4'],
-                'joinsource_name_or_index': 'F4',
-                'label': 'Label',
-                'label_values_or_threshold': [1],
-                'methods': {
-                    'report': {'name': 'report', 'title': 'Analysis Report'},
-                    'post_training_bias': {'methods': 'all'},
-                    'pre_training_bias': {'methods': 'all'}},
-                'predictor': {'initial_instance_count': 1,
-                              'instance_type': 'ml.c5.xlarge',
-                              'label_headers': ['success'],
-                              'model_name': 'xgboost-model',
-                              'probability': 'pr'}}
+    expected = {
+        "dataset_type": "text/csv",
+        "facet": [{"name_or_index": "F1"}],
+        "group_variable": "F2",
+        "headers": ["Label", "F1", "F2", "F3", "F4"],
+        "joinsource_name_or_index": "F4",
+        "label": "Label",
+        "label_values_or_threshold": [1],
+        "methods": {
+            "report": {"name": "report", "title": "Analysis Report"},
+            "post_training_bias": {"methods": "all"},
+            "pre_training_bias": {"methods": "all"},
+        },
+        "predictor": {
+            "initial_instance_count": 1,
+            "instance_type": "ml.c5.xlarge",
+            "label_headers": ["success"],
+            "model_name": "xgboost-model",
+            "probability": "pr",
+        },
+    }
     assert actual == expected