Update Catboost tuning

bluecast/blueprints/cast.py

@@ -15,6 +15,8 @@
 import pandas as pd
 
 from bluecast.config.training_config import (
+    CatboostFinalParamConfig,
+    CatboostTuneParamsConfig,
     TrainingConfig,
     XgboostFinalParamConfig,
     XgboostTuneParamsConfig,
@@ -97,8 +99,12 @@ def __init__(
             Union[BoostaRootaWrapper, CustomPreprocessing]
         ] = None,
         conf_training: Optional[TrainingConfig] = None,
-        conf_xgboost: Optional[XgboostTuneParamsConfig] = None,
-        conf_params_xgboost: Optional[XgboostFinalParamConfig] = None,
+        conf_xgboost: Optional[
+            Union[XgboostTuneParamsConfig, CatboostTuneParamsConfig]
+        ] = None,
+        conf_params_xgboost: Optional[
+            Union[XgboostFinalParamConfig, CatboostFinalParamConfig]
+        ] = None,
         experiment_tracker: Optional[ExperimentTracker] = None,
         single_fold_eval_metric_func: Optional[ClassificationEvalWrapper] = None,
     ):

bluecast/blueprints/cast_cv.py

@@ -8,6 +8,8 @@
 
 from bluecast.blueprints.cast import BlueCast
 from bluecast.config.training_config import (
+    CatboostFinalParamConfig,
+    CatboostTuneParamsConfig,
     TrainingConfig,
     XgboostFinalParamConfig,
     XgboostTuneParamsConfig,
@@ -58,8 +60,12 @@ def __init__(
         cat_columns: Optional[List[Union[str, float, int]]] = None,
         stratifier: Optional[Any] = None,
         conf_training: Optional[TrainingConfig] = None,
-        conf_xgboost: Optional[XgboostTuneParamsConfig] = None,
-        conf_params_xgboost: Optional[XgboostFinalParamConfig] = None,
+        conf_xgboost: Optional[
+            Union[XgboostTuneParamsConfig, CatboostTuneParamsConfig]
+        ] = None,
+        conf_params_xgboost: Optional[
+            Union[XgboostFinalParamConfig, CatboostFinalParamConfig]
+        ] = None,
         experiment_tracker: Optional[ExperimentTracker] = None,
         custom_in_fold_preprocessor: Optional[CustomPreprocessing] = None,
         custom_last_mile_computation: Optional[CustomPreprocessing] = None,

bluecast/blueprints/cast_cv_regression.py

@@ -10,6 +10,8 @@
 
 from bluecast.blueprints.cast_regression import BlueCastRegression
 from bluecast.config.training_config import (
+    CatboostRegressionFinalParamConfig,
+    CatboostTuneParamsRegressionConfig,
     TrainingConfig,
     XgboostRegressionFinalParamConfig,
     XgboostTuneParamsRegressionConfig,
@@ -61,8 +63,12 @@ def __init__(
         cat_columns: Optional[List[Union[str, float, int]]] = None,
         stratifier: Optional[Any] = None,
         conf_training: Optional[TrainingConfig] = None,
-        conf_xgboost: Optional[XgboostTuneParamsRegressionConfig] = None,
-        conf_params_xgboost: Optional[XgboostRegressionFinalParamConfig] = None,
+        conf_xgboost: Optional[
+            Union[XgboostTuneParamsRegressionConfig, CatboostTuneParamsRegressionConfig]
+        ] = None,
+        conf_params_xgboost: Optional[
+            Union[XgboostRegressionFinalParamConfig, CatboostRegressionFinalParamConfig]
+        ] = None,
         experiment_tracker: Optional[ExperimentTracker] = None,
         custom_in_fold_preprocessor: Optional[CustomPreprocessing] = None,
         custom_last_mile_computation: Optional[CustomPreprocessing] = None,

bluecast/blueprints/cast_regression.py

@@ -16,6 +16,8 @@
 from sklearn.metrics import mean_squared_error
 
 from bluecast.config.training_config import (
+    CatboostRegressionFinalParamConfig,
+    CatboostTuneParamsRegressionConfig,
     TrainingConfig,
     XgboostRegressionFinalParamConfig,
     XgboostTuneParamsRegressionConfig,
@@ -93,8 +95,12 @@ def __init__(
             Union[BoostaRootaWrapper, CustomPreprocessing]
         ] = None,
         conf_training: Optional[TrainingConfig] = None,
-        conf_xgboost: Optional[XgboostTuneParamsRegressionConfig] = None,
-        conf_params_xgboost: Optional[XgboostRegressionFinalParamConfig] = None,
+        conf_xgboost: Optional[
+            Union[XgboostTuneParamsRegressionConfig, CatboostTuneParamsRegressionConfig]
+        ] = None,
+        conf_params_xgboost: Optional[
+            Union[XgboostRegressionFinalParamConfig, CatboostRegressionFinalParamConfig]
+        ] = None,
         experiment_tracker: Optional[ExperimentTracker] = None,
         single_fold_eval_metric_func: Optional[RegressionEvalWrapper] = None,
     ):

bluecast/config/training_config.py

@@ -516,7 +516,8 @@ def __init__(
     ):
         if bootstrap_type is None:
             bootstrap_type = [
-                "Bayesian"
+                "Bayesian",
+                "No",
             ]  # Poisson not possible on CPU, "MVS" requires min samples
         if grow_policy is None:
             grow_policy = ["SymmetricTree"]
@@ -615,7 +616,7 @@ def __init__(
         catboost_eval_metric_tune_direction: str = "minimize",
     ):
         if bootstrap_type is None:
-            bootstrap_type = ["Bayesian", "Poisson", "MVS"]
+            bootstrap_type = ["Bayesian", "No"]  # "Poisson", "MVS"
         if grow_policy is None:
             grow_policy = ["SymmetricTree"]
 

bluecast/ml_modelling/catboost.py

@@ -227,7 +227,7 @@ def objective(trial):
                     log=True,
                 ),
             }
-            if params["bootstrap_type"] == "Bayesian":
+            if params["bootstrap_type"] in ["Bayesian", "No"]:
                 params["bagging_temperature"] = None
                 params["subsample"] = None
 
@@ -383,7 +383,7 @@ def objective(trial):
                 }
                 final_best_params = {**final_best_params, **train_on}
 
-                if final_best_params["bootstrap_type"] == "Bayesian":
+                if final_best_params["bootstrap_type"] in ["Bayesian", "No"]:
                     final_best_params.pop("subsample", None)
                     final_best_params.pop("bagging_temperature", None)
 

bluecast/ml_modelling/catboost_regression.py

@@ -223,7 +223,7 @@ def objective(trial):
                 ),
             }
 
-            if params["bootstrap_type"] == "Bayesian":
+            if params["bootstrap_type"] in ["Bayesian", "No"]:
                 params["bagging_temperature"] = None
                 params["subsample"] = None
 
@@ -359,7 +359,7 @@ def objective(trial):
                 # Merge device or other settings
                 final_best_params = {**final_best_params, **train_on}
 
-                if final_best_params["bootstrap_type"] == "Bayesian":
+                if final_best_params["bootstrap_type"] in ["Bayesian", "No"]:
                     final_best_params.pop("subsample", None)
                     final_best_params.pop("bagging_temperature", None)
 

bluecast/tests/test_catboost_regression.py

@@ -0,0 +1,129 @@
+import numpy as np
+import pandas as pd
+
+from bluecast.blueprints.cast_regression import BlueCastRegression
+from bluecast.config.training_config import (
+    CatboostTuneParamsRegressionConfig,
+    TrainingConfig,
+)
+from bluecast.ml_modelling.catboost_regression import CatboostModelRegression
+
+
+def test_BlueCastRegression_without_hyperparam_tuning():
+    train_config = TrainingConfig()
+    train_config.hyperparameter_tuning_rounds = 10
+    train_config.hypertuning_cv_folds = 2
+    train_config.autotune_model = False
+
+    catboost_pram_config = CatboostTuneParamsRegressionConfig()
+
+    # Create an instance of the BlueCastRegression class with the custom model
+    bluecast = BlueCastRegression(
+        class_problem="binary",
+        ml_model=CatboostModelRegression(
+            class_problem="regression",
+            conf_training=train_config,
+            conf_catboost=catboost_pram_config,
+        ),
+        conf_xgboost=catboost_pram_config,
+        conf_training=train_config,
+    )
+
+    # Create some sample data for testing
+    x_train = pd.DataFrame(
+        {
+            "feature1": [i for i in range(20)],
+            "feature2": [i for i in range(20)],
+            "feature3": [i for i in range(20)],
+            "feature4": [i for i in range(20)],
+            "feature5": [i for i in range(20)],
+            "feature6": [i for i in range(20)],
+        }
+    )
+    y_train = pd.Series([0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1])
+    x_test = pd.DataFrame(
+        {
+            "feature1": [i for i in range(10)],
+            "feature2": [i for i in range(10)],
+            "feature3": [i for i in range(10)],
+            "feature4": [i for i in range(10)],
+            "feature5": [i for i in range(10)],
+            "feature6": [i for i in range(10)],
+        }
+    )
+
+    x_train["target"] = y_train
+
+    # Fit the BlueCastRegression model using the custom model
+    bluecast.fit(x_train, "target")
+
+    # Predict on the test data using the custom model
+    predicted_probas, predicted_classes = bluecast.predict(x_test)
+
+    # Assert the expected results
+    assert isinstance(predicted_probas, np.ndarray)
+    assert isinstance(predicted_classes, np.ndarray)
+    print(bluecast.experiment_tracker.experiment_id)
+    assert (
+        len(bluecast.experiment_tracker.experiment_id) == 0
+    )  # due to custom model and fit method
+
+
+def test_BlueCastRegression_with_hyperparam_tuning():
+    train_config = TrainingConfig()
+    train_config.hyperparameter_tuning_rounds = 10
+    train_config.hypertuning_cv_folds = 2
+    train_config.autotune_model = True
+
+    catboost_pram_config = CatboostTuneParamsRegressionConfig()
+
+    # Create an instance of the BlueCastRegression class with the custom model
+    bluecast = BlueCastRegression(
+        class_problem="regression",
+        ml_model=CatboostModelRegression(
+            class_problem="regression",
+            conf_training=train_config,
+            conf_catboost=catboost_pram_config,
+        ),
+        conf_xgboost=catboost_pram_config,
+        conf_training=train_config,
+    )
+
+    # Create some sample data for testing
+    x_train = pd.DataFrame(
+        {
+            "feature1": [i for i in range(20)],
+            "feature2": [i for i in range(20)],
+            "feature3": [i for i in range(20)],
+            "feature4": [i for i in range(20)],
+            "feature5": [i for i in range(20)],
+            "feature6": [i for i in range(20)],
+        }
+    )
+    y_train = pd.Series([0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1])
+    x_test = pd.DataFrame(
+        {
+            "feature1": [i for i in range(10)],
+            "feature2": [i for i in range(10)],
+            "feature3": [i for i in range(10)],
+            "feature4": [i for i in range(10)],
+            "feature5": [i for i in range(10)],
+            "feature6": [i for i in range(10)],
+        }
+    )
+
+    x_train["target"] = y_train
+
+    # Fit the BlueCastRegression model using the custom model
+    bluecast.fit(x_train, "target")
+
+    # Predict on the test data using the custom model
+    predicted_probas, predicted_classes = bluecast.predict(x_test)
+
+    # Assert the expected results
+    assert isinstance(predicted_probas, np.ndarray)
+    assert isinstance(predicted_classes, np.ndarray)
+    print(bluecast.experiment_tracker.experiment_id)
+    assert (
+        len(bluecast.experiment_tracker.experiment_id) == 0
+    )  # due to custom model and fit method