teradataml 20.0.0.0__py3-none-any.whl → 20.0.0.1__py3-none-any.whl

teradataml/automl/__init__.py

@@ -30,6 +30,7 @@ from teradataml.utils.dtypes import _Dtypes
 from teradataml.common.utils import UtilFuncs
 from teradataml import TeradataMlException
 from teradataml.common.messages import Messages, MessageCodes
+from teradatasqlalchemy.telemetry.queryband import collect_queryband
 
 # AutoML Internal libraries
 from teradataml.automl.data_preparation import _DataPreparation
@@ -51,6 +52,7 @@ class AutoML:
                  max_runtime_secs = None,
                  stopping_metric = None, 
                  stopping_tolerance = None,
+                 max_models = None,
                  custom_config_file = None):
         """
         DESCRIPTION:
@@ -82,12 +84,12 @@ class AutoML:
             AutoML also provides an option to customize the processes within feature
             engineering, data preparation and model training phases. User can customize
             the processes by passing the JSON file path in case of custom run. It also 
-            supports early stopping of model training based on stopping metrics and
-            maximum running time.
+            supports early stopping of model training based on stopping metrics,
+            maximum running time and maximum models to be trained.
          
         PARAMETERS:  
             task_type:
-                Optional Arugment.
+                Optional Argument.
                 Specifies the task type for AutoML, whether to apply regression OR classification
                 on the provided dataset. If user wants AutoML to decide the task type automatically, 
                 then it should be set to "Default".
@@ -122,7 +124,7 @@ class AutoML:
                 Types: int
                 
             max_runtime_secs:
-                Optional Arugment.
+                Optional Argument.
                 Specifies the time limit in seconds for model training.
                 Types: int
                 
@@ -143,6 +145,11 @@ class AutoML:
                 Required, when "stopping_metric" is set, otherwise optional.
                 Specifies the stopping tolerance for stopping metrics in model training.
                 Types: float
+            
+            max_models:
+                Optional Argument.
+                Specifies the maximum number of models to be trained.
+                Types: int
                 
             custom_config_file:
                 Optional Argument.
@@ -285,6 +292,7 @@ class AutoML:
             >>>                     exclude="xgboost",
             >>>                     stopping_metric="R2",
             >>>                     stopping_tolerance=0.7,
+            >>>                     max_models=10,
             >>>                     custom_config_file="custom_housing.json")
             # Fit the data.
             >>> automl_obj.fit(housing_train, "price")
@@ -303,7 +311,8 @@ class AutoML:
             # Create instance of AutoML.
             >>> automl_obj = AutoML(verbose=2, 
             >>>                     exclude="xgboost",
-            >>>                     max_runtime_secs=500)
+            >>>                     max_runtime_secs=500,
+            >>>                     max_models=3)
             # Fit the data.
             >>> automl_obj.fit(iris_input, iris_input.species)
  
@@ -339,6 +348,7 @@ class AutoML:
                                                                                         'WEIGHTED-PRECISION','WEIGHTED-RECALL',
                                                                                         'WEIGHTED-F1', 'ACCURACY']])
         arg_info_matrix.append(["stopping_tolerance", stopping_tolerance, True, (float, int)])
+        arg_info_matrix.append(["max_models", max_models, True, (int)])
         arg_info_matrix.append(["custom_config_file", custom_config_file, True, (str), True])
         
 
@@ -349,6 +359,8 @@ class AutoML:
             _Validators._validate_mutually_exclusive_arguments(include, "include", exclude, "exclude")
         # Validate mutually inclusive arguments
         _Validators._validate_mutually_inclusive_arguments(stopping_metric, "stopping_metric", stopping_tolerance, "stopping_tolerance")
+        # Validate lower range for max_models
+        _Validators._validate_argument_range(max_models, "max_models", lbound=1, lbound_inclusive=True)
         
         custom_data = None
         self.auto = True
@@ -375,10 +387,12 @@ class AutoML:
         self.max_runtime_secs = max_runtime_secs
         self.stopping_metric = stopping_metric
         self.stopping_tolerance = stopping_tolerance
+        self.max_models = max_models
         self.model_list = ['decision_forest', 'xgboost', 'knn', 'svm', 'glm']
         self.is_classification_type = lambda: self.task_type.upper() == 'CLASSIFICATION'
         self._is_fit_called = False 
         
+    @collect_queryband(queryband="AutoML_fit")    
     def fit(self,
             data,
             target_column):
@@ -394,7 +408,7 @@ class AutoML:
                 Types: teradataml Dataframe
             
             target_column:
-                Required Arugment.
+                Required Argument.
                 Specifies target column of dataset.
                 Types: str or ColumnExpression
 
@@ -484,7 +498,7 @@ class AutoML:
         
         # Displaying received custom input
         if self.custom_data:
-            print("\n Received below input for customization : ")
+            print("\nReceived below input for customization : ")
             print(json.dumps(self.custom_data, indent=4))
         
         # Classification probelm
@@ -506,13 +520,15 @@ class AutoML:
                                                                     verbose = self.verbose,
                                                                     max_runtime_secs = self.max_runtime_secs, 
                                                                     stopping_metric = self.stopping_metric, 
-                                                                    stopping_tolerance = self.stopping_tolerance
-                                                                    )
+                                                                    stopping_tolerance = self.stopping_tolerance,
+                                                                    max_models = self.max_models)
+
         # Model Evaluation Phase
         self.m_evaluator = _ModelEvaluator(self.model_info, 
                                            self.target_column,
                                            self.task_type)
       
+    @collect_queryband(queryband="AutoML_predict")  
     def predict(self,
                 data = None,
                 rank = 1):
@@ -582,6 +598,10 @@ class AutoML:
 
         # Validate argument types
         _Validators._validate_function_arguments(arg_info_pred_matrix)
+        # Validate range for model rank
+        _Validators._validate_argument_range(rank, "rank", lbound=1, 
+                                             ubound=self.leader_board.Rank.max(),
+                                             lbound_inclusive=True, ubound_inclusive=True)
         
         # Setting test data indicator to default value, i.e., False.
         self.test_data_ind = False
@@ -633,7 +653,7 @@ class AutoML:
         if self.is_classification_type() and self.target_label is not None:
             # Displaying target column labels
             tar_dct = {}
-            print('Target Column Mapping:')
+            print('\nTarget Column Mapping:')
             # Iterating rows
             for row in self.target_label.result.itertuples():
                 # Retrieving the category names of encoded target column
@@ -645,13 +665,13 @@ class AutoML:
             for key, value in tar_dct.items():
                 print(f"{key}: {value}")
             
-        print("\n Prediction : ")
+        print("\nPrediction : ")
         print(pred.result)
         
         # Showing performance metrics if there is no test data
         # Or if target column is present in test data.
         if not self.test_data_ind or self.target_column_ind:
-            print("\n Performance Metrics : ")
+            print("\nPerformance Metrics : ")
             print(metrics.result)
         
             prediction_column = 'prediction' if 'prediction' in pred.result.columns else 'Prediction'
@@ -669,19 +689,20 @@ class AutoML:
                     }
                     # Fitting ROC
                     roc_out = ROC(**fit_params)
-                    print("\n ROC-AUC : ")
+                    print("\nROC-AUC : ")
                     print_data(roc_out.result)
                     print_data(roc_out.output_data)
                 
                 # Displaying confusion matrix for binary and multiclass classification
                 prediction_df=pred.result.to_pandas()
                 target_col = self.target_column
-                print("\n Confusion Matrix : ")
+                print("\nConfusion Matrix : ")
                 print_data(confusion_matrix(prediction_df[target_col], prediction_df[prediction_column]))
                 
         # Returning prediction
         return pred.result    
     
+    @collect_queryband(queryband="AutoML_leaderboard")
     def leaderboard(self):
         """
         DESCRIPTION:
@@ -709,6 +730,7 @@ class AutoML:
             raise TeradataMlException(err, MessageCodes.EXECUTION_FAILED)
         return self.leader_board
         
+    @collect_queryband(queryband="AutoML_leader")    
     def leader(self):
         """
         DESCRIPTION:
@@ -810,12 +832,12 @@ class _Regression(_FeatureExplore, _FeatureEngineering, _DataPreparation, _Model
                 Types: teradataml Dataframe
             
             target_column:
-                Required Arugment.
+                Required Argument.
                 Specifies the name of the target column in "data".
                 Types: str
             
             custom_data:
-                Optional Arugment.
+                Optional Argument.
                 Specifies json object containing user customized input.
                 Types: json object
         """
@@ -830,14 +852,15 @@ class _Regression(_FeatureExplore, _FeatureEngineering, _DataPreparation, _Model
                     verbose = 0,
                     max_runtime_secs = None,
                     stopping_metric = None, 
-                    stopping_tolerance = None):
+                    stopping_tolerance = None,
+                    max_models = None):
         """
         DESCRIPTION:
             Interal Function runs Regression.
          
         PARAMETERS:  
             auto:
-                Optional Arugment.
+                Optional Argument.
                 Specifies whether to run AutoML in custom mode or auto mode.
                 When set to False, runs in custom mode. Otherwise, by default runs in auto mode.
                 Types: bool
@@ -853,19 +876,24 @@ class _Regression(_FeatureExplore, _FeatureEngineering, _DataPreparation, _Model
                 Types: int
                 
             max_runtime_secs:
-                Optional Arugment.
+                Optional Argument.
                 Specifies the time limit in seconds for model training.
                 Types: int
                 
             stopping_metric:
                 Required, when "stopping_tolerance" is set, otherwise optional.
-	        Specifies the stopping mertics for stopping tolerance in model training.
+                Specifies the stopping mertics for stopping tolerance in model training.
                 Types: str
 
             stopping_tolerance:
                 Required, when "stopping_metric" is set, otherwise optional.
-	        Specifies the stopping tolerance for stopping metrics in model training.
+                Specifies the stopping tolerance for stopping metrics in model training.
                 Types: float
+            
+            max_models:
+                Optional Argument.
+                Specifies the maximum number of models to be trained.
+                Types: int
         
         RETURNS:
             a tuple containing, model information and leaderboard.   
@@ -919,7 +947,8 @@ class _Regression(_FeatureExplore, _FeatureEngineering, _DataPreparation, _Model
         models_info, leaderboard, target_count = self.model_training(auto = auto,
                                                         max_runtime_secs = max_runtime_secs, 
                                                         stopping_metric = stopping_metric, 
-                                                        stopping_tolerance = stopping_tolerance)
+                                                        stopping_tolerance = stopping_tolerance,
+                                                        max_models = max_models)
 
         return (models_info, leaderboard, target_count, target_label, data_transformation_params, self.table_name_mapping)
             
@@ -940,12 +969,12 @@ class _Classification(_FeatureExplore, _FeatureEngineering, _DataPreparation, _M
                 Types: teradataml Dataframe
             
             target_column:
-                Required Arugment.
+                Required Argument.
                 Specifies the name of the target column in "data".
                 Types: str
             
             custom_data:
-                Optional Arugment.
+                Optional Argument.
                 Specifies json object containing user customized input.
                 Types: json object
         """
@@ -959,14 +988,15 @@ class _Classification(_FeatureExplore, _FeatureEngineering, _DataPreparation, _M
                         verbose = 0,
                         max_runtime_secs = None,
                         stopping_metric = None, 
-                        stopping_tolerance = None):
+                        stopping_tolerance = None,
+                        max_models = None):
         """
         DESCRIPTION:
             Interal Function runs Classification.
          
         PARAMETERS:  
             auto:
-                Optional Arugment.
+                Optional Argument.
                 Specifies whether to run AutoML in custom mode or auto mode.
                 When set to False, runs in custom mode. Otherwise, by default runs in auto mode.
                 Types: bool
@@ -982,7 +1012,7 @@ class _Classification(_FeatureExplore, _FeatureEngineering, _DataPreparation, _M
                 Types: int
                 
             max_runtime_secs:
-                Optional Arugment.
+                Optional Argument.
                 Specifies the time limit in seconds for model training.
                 Types: int
                 
@@ -995,6 +1025,11 @@ class _Classification(_FeatureExplore, _FeatureEngineering, _DataPreparation, _M
                 Required, when "stopping_metric" is set, otherwise optional.
 	            Specifies the stopping tolerance for stopping metrics in model training.
                 Types: float
+            
+            max_models:
+                Optional Argument.
+                Specifies the maximum number of models to be trained.
+                Types: int
         
         RETURNS:
             a tuple containing, model information and leaderboard.
@@ -1051,7 +1086,8 @@ class _Classification(_FeatureExplore, _FeatureEngineering, _DataPreparation, _M
         models_info, leaderboard, target_count = self.model_training(auto = auto,
                                                         max_runtime_secs = max_runtime_secs, 
                                                         stopping_metric = stopping_metric, 
-                                                        stopping_tolerance = stopping_tolerance)
+                                                        stopping_tolerance = stopping_tolerance,
+                                                        max_models = max_models)
 
         return (models_info, leaderboard, target_count, target_label, data_transformation_params, self.table_name_mapping)
   
@@ -1166,7 +1202,7 @@ class _Classification(_FeatureExplore, _FeatureEngineering, _DataPreparation, _M
             min_label_count = min(data[self.target_column].value_counts())
             if self._data_sampling_method == 'SMOTE':
                 n_neighbors = min(5, min_label_count - 1)
-                sampling_method = SMOTE(k_neighbors=n_neighbors, random_state=5)
+                sampling_method = SMOTE(k_neighbors=n_neighbors, random_state=42)
             else:
                 n_neighbors = min(3, min_label_count)
                 sampling_method = NearMiss(version=1, n_neighbors=n_neighbors)
@@ -1206,6 +1242,7 @@ class AutoRegressor(AutoML):
                  max_runtime_secs=None,
                  stopping_metric=None, 
                  stopping_tolerance=None,
+                 max_models=None,
                  custom_config_file=None
                 ):
         """
@@ -1239,7 +1276,7 @@ class AutoRegressor(AutoML):
                 Types: int
                 
             max_runtime_secs:
-                Optional Arugment.
+                Optional Argument.
                 Specifies the time limit in seconds for model training.
                 Types: int
                 
@@ -1260,6 +1297,11 @@ class AutoRegressor(AutoML):
                 Required, when "stopping_metric" is set, otherwise optional.
                 Specifies the stopping tolerance for stopping metrics in model training.
                 Types: float
+            
+            max_models:
+                Optional Argument.
+                Specifies the maximum number of models to be trained.
+                Types: int
                 
             custom_config_file:
                 Optional Argument.
@@ -1325,10 +1367,11 @@ class AutoRegressor(AutoML):
 
             # Create instance of AutoRegressor.
             >>> automl_obj = AutoRegressor(verbose=2,
-            >>>                     exclude="xgboost",
-            >>>                     stopping_metric="R2",
-            >>>                     stopping_tolerance=0.7,
-            >>>                     custom_config_file="custom_housing.json")
+            >>>                            exclude="xgboost",
+            >>>                            stopping_metric="R2",
+            >>>                            stopping_tolerance=0.7,
+            >>>                            max_models=10,
+            >>>                            custom_config_file="custom_housing.json")
             # Fit the data.
             >>> automl_obj.fit(housing_train, "price")
  
@@ -1345,8 +1388,8 @@ class AutoRegressor(AutoML):
 
             # Create instance of AutoRegressor.
             >>> automl_obj = AutoRegressor(verbose=2, 
-            >>>                     exclude="xgboost",
-            >>>                     max_runtime_secs=500)
+            >>>                            exclude="xgboost",
+            >>>                            max_runtime_secs=500)
             # Fit the data.
             >>> automl_obj.fit(housing_train, "price")
  
@@ -1368,6 +1411,7 @@ class AutoRegressor(AutoML):
         self.max_runtime_secs = max_runtime_secs
         self.stopping_metric = stopping_metric
         self.stopping_tolerance = stopping_tolerance
+        self.max_models = max_models
         self.custom_config_file = custom_config_file
         self.task_type = "Regression"
         self.include = include
@@ -1380,6 +1424,7 @@ class AutoRegressor(AutoML):
                                             max_runtime_secs=self.max_runtime_secs,
                                             stopping_metric=self.stopping_metric, 
                                             stopping_tolerance=self.stopping_tolerance,
+                                            max_models=self.max_models,
                                             custom_config_file=self.custom_config_file)        
 class AutoClassifier(AutoML):
         
@@ -1390,6 +1435,7 @@ class AutoClassifier(AutoML):
                 max_runtime_secs=None,
                 stopping_metric=None, 
                 stopping_tolerance=None,
+                max_models=None,
                 custom_config_file=None
                 ):
         """
@@ -1423,18 +1469,13 @@ class AutoClassifier(AutoML):
                 Types: int
                 
             max_runtime_secs:
-                Optional Arugment.
+                Optional Argument.
                 Specifies the time limit in seconds for model training.
                 Types: int
                 
             stopping_metric:
                 Required, when "stopping_tolerance" is set, otherwise optional.
                 Specifies the stopping mertics for stopping tolerance in model training.
-                Types: str
-
-            stopping_tolerance:
-                Required, when "stopping_metric" is set, otherwise optional.
-                Specifies the stopping tolerance for stopping metrics in model training.
                 Permitted Values: 
                     * For task_type "Regression": "R2", "MAE", "MSE", "MSLE", 
                                                   "RMSE", "RMSLE"
@@ -1443,7 +1484,17 @@ class AutoClassifier(AutoML):
                                                       'MICRO-PRECISION', 'MACRO-PRECISION',
                                                       'WEIGHTED-PRECISION','WEIGHTED-RECALL',
                                                       'WEIGHTED-F1', 'ACCURACY'
+                Types: str
+
+            stopping_tolerance:
+                Required, when "stopping_metric" is set, otherwise optional.
+                Specifies the stopping tolerance for stopping metrics in model training.
                 Types: float
+            
+            max_models:
+                Optional Argument.
+                Specifies the maximum number of models to be trained.
+                Types: int
                 
             custom_config_file:
                 Optional Argument.
@@ -1570,10 +1621,11 @@ class AutoClassifier(AutoML):
 
             # Create instance of AutoClassifier.
             >>> automl_obj = AutoClassifier(verbose=2, 
-            >>>                     exclude="xgboost",
-            >>>                     stopping_metric="MICRO-F1",
-            >>>                     stopping_tolerance=0.7,
-            >>>                     custom_config_file="custom_titanic.json")
+            >>>                             exclude="xgboost",
+            >>>                             stopping_metric="MICRO-F1",
+            >>>                             stopping_tolerance=0.7,
+            >>>                             max_models=8
+            >>>                             custom_config_file="custom_titanic.json")
             # Fit the data.
             >>> automl_obj.fit(titanic, titanic.survived)
  
@@ -1590,8 +1642,9 @@ class AutoClassifier(AutoML):
 
             # Create instance of AutoClassifier.
             >>> automl_obj = AutoClassifier(verbose=2, 
-            >>>                     exclude="xgboost",
-            >>>                     max_runtime_secs=500)
+            >>>                             exclude="xgboost",
+            >>>                             max_runtime_secs=500)
+            >>>                             max_models=3)
             # Fit the data.
             >>> automl_obj.fit(iris_input, iris_input.species)
  
@@ -1613,6 +1666,7 @@ class AutoClassifier(AutoML):
         self.max_runtime_secs = max_runtime_secs
         self.stopping_metric = stopping_metric
         self.stopping_tolerance = stopping_tolerance
+        self.max_models = max_models
         self.custom_config_file = custom_config_file
         self.task_type = "Classification"
         self.include = include
@@ -1625,4 +1679,5 @@ class AutoClassifier(AutoML):
                                             max_runtime_secs=self.max_runtime_secs,
                                             stopping_metric=self.stopping_metric, 
                                             stopping_tolerance=self.stopping_tolerance,
+                                            max_models=self.max_models,
                                             custom_config_file=self.custom_config_file)