Merge pull request #128 from perib/new_search_space_def

New search space def

tpot2/config/classifiers.py

@@ -3,7 +3,7 @@
 from ConfigSpace import EqualsCondition, OrConjunction, NotEqualsCondition, InCondition
 from ..search_spaces.nodes.estimator_node import NONE_SPECIAL_STRING, TRUE_SPECIAL_STRING, FALSE_SPECIAL_STRING
 import numpy as np
-
+import sklearn
 
 
 #TODO Conditional search space to prevent invalid combinations of hyperparameters
@@ -438,4 +438,42 @@ def MLPClassifier_hyperparameter_parser(params):
         'learning_rate': params['learning_rate'],
         'early_stopping': params['early_stopping'],
     }
-    return hyperparameters
+    return hyperparameters
+
+###
+
+
+###
+
+def get_GaussianProcessClassifier_ConfigurationSpace(n_features, random_state):
+    space = {
+        'n_features': n_features,
+        'alpha': Float("alpha", bounds=(1e-14, 1.0), log=True),
+        'thetaL': Float("thetaL", bounds=(1e-10, 1e-3), log=True),
+        'thetaU': Float("thetaU", bounds=(1.0, 100000), log=True),
+    }
+
+    if random_state is not None: #This is required because configspace doesn't allow None as a value
+        space['random_state'] = random_state
+
+    return ConfigurationSpace(
+        space = space
+    )
+
+def GaussianProcessClassifier_hyperparameter_parser(params):
+    kernel = sklearn.gaussian_process.kernels.RBF(
+        length_scale = [1.0]*params['n_features'],
+        length_scale_bounds=[(params['thetaL'], params['thetaU'])] * params['n_features'],
+    )
+    final_params = {"kernel": kernel, 
+                    "alpha": params['alpha'],
+                    "n_restarts_optimizer": 10,
+                    "optimizer": "fmin_l_bfgs_b",
+                    "normalize_y": True,
+                    "copy_X_train": True,
+                    }
+
+    if "random_state" in params:
+        final_params['random_state'] = params['random_state']
+
+    return final_params

tpot2/config/get_configspace.py

@@ -92,9 +92,8 @@
 from sklearn.linear_model import PassiveAggressiveClassifier
 from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis
 from sklearn.linear_model import ARDRegression
-from sklearn.gaussian_process import GaussianProcessRegressor
+from sklearn.gaussian_process import GaussianProcessRegressor, GaussianProcessClassifier
 
-from sklearn.gaussian_process import GaussianProcessRegressor
 
 from xgboost import XGBRegressor
 
@@ -113,6 +112,7 @@
                AddTransformer, mul_neg_1_Transformer, MulTransformer, SafeReciprocalTransformer, EQTransformer, NETransformer, GETransformer, GTTransformer, LETransformer, LTTransformer, MinTransformer, MaxTransformer, ZeroTransformer, OneTransformer, NTransformer,
                PowerTransformer, QuantileTransformer,ARDRegression, QuadraticDiscriminantAnalysis, PassiveAggressiveClassifier, LinearDiscriminantAnalysis,
                DominantEncoder, RecessiveEncoder, HeterosisEncoder, UnderDominanceEncoder, OverDominanceEncoder,
+               GaussianProcessClassifier,
                ]
 
 
@@ -168,7 +168,7 @@
         "selectors": ["SelectFwe", "SelectPercentile", "VarianceThreshold",],
         "selectors_classification": ["SelectFwe", "SelectPercentile", "VarianceThreshold", "RFE_classification", "SelectFromModel_classification"],
         "selectors_regression": ["SelectFwe", "SelectPercentile", "VarianceThreshold", "RFE_regression", "SelectFromModel_regression"],
-        "classifiers" :  ['AdaBoostClassifier', 'BernoulliNB', 'DecisionTreeClassifier', 'ExtraTreesClassifier', 'GaussianNB', 'HistGradientBoostingClassifier', 'KNeighborsClassifier','LinearDiscriminantAnalysis', 'LogisticRegression', "LinearSVC", "SVC", 'MLPClassifier', 'MultinomialNB', "PassiveAggressiveClassifier", "QuadraticDiscriminantAnalysis", 'RandomForestClassifier', 'SGDClassifier', 'XGBClassifier'],
+        "classifiers" :  ["GaussianProcessClassifier", 'AdaBoostClassifier', 'BernoulliNB', 'DecisionTreeClassifier', 'ExtraTreesClassifier', 'GaussianNB', 'HistGradientBoostingClassifier', 'KNeighborsClassifier','LinearDiscriminantAnalysis', 'LogisticRegression', "LinearSVC", "SVC", 'MLPClassifier', 'MultinomialNB', "PassiveAggressiveClassifier", "QuadraticDiscriminantAnalysis", 'RandomForestClassifier', 'SGDClassifier', 'XGBClassifier'],
         "regressors" : ['AdaBoostRegressor', "ARDRegression", 'DecisionTreeRegressor', 'ExtraTreesRegressor', 'GaussianProcessRegressor', 'HistGradientBoostingRegressor', 'KNeighborsRegressor',  'LinearSVR', "MLPRegressor", 'RandomForestRegressor', 'SGDRegressor', 'SVR', 'XGBRegressor'],
 
 
@@ -252,6 +252,8 @@ def get_configspace(name, n_classes=3, n_samples=100, n_features=100, random_sta
             return classifiers.get_PassiveAggressiveClassifier_ConfigurationSpace(random_state=random_state)
         case "QuadraticDiscriminantAnalysis":
             return classifiers.get_QuadraticDiscriminantAnalysis_ConfigurationSpace()
+        case "GaussianProcessClassifier":
+            return classifiers.get_GaussianProcessClassifier_ConfigurationSpace(n_features=n_features, random_state=random_state)
 
         #regressors.py
         case "RandomForestRegressor":
@@ -498,11 +500,14 @@ def get_node(name, n_classes=3, n_samples=100, n_features=100, random_state=None
     if name == "GaussianProcessRegressor":
         configspace = get_configspace(name, n_classes=n_classes, n_samples=n_samples, random_state=random_state)
         return EstimatorNode(STRING_TO_CLASS[name], configspace, hyperparameter_parser=regressors.GaussianProcessRegressor_hyperparameter_parser)
+    if name == "GaussianProcessClassifier":
+        configspace = get_configspace(name, n_classes=n_classes, n_samples=n_samples, random_state=random_state)
+        return EstimatorNode(STRING_TO_CLASS[name], configspace, hyperparameter_parser=classifiers.GaussianProcessClassifier_hyperparameter_parser)
 
     configspace = get_configspace(name, n_classes=n_classes, n_samples=n_samples, n_features=n_features, random_state=random_state)
     if configspace is None:
         #raise warning
         warnings.warn(f"Could not find configspace for {name}")
         return None
 
-    return EstimatorNode(STRING_TO_CLASS[name], configspace)
+    return EstimatorNode(STRING_TO_CLASS[name], configspace)