add radial basis functions

autoemulate/emulators/__init__.py

@@ -3,10 +3,12 @@
 from .gaussian_process_sk import GaussianProcessSk
 from .neural_network import NeuralNetwork
 from .random_forest import RandomForest
+from .radial_basis import RadialBasis
 
 MODEL_REGISTRY = {
     # "GaussianProcess": GaussianProcess,
     "GaussianProcessSk": GaussianProcessSk,
     "NeuralNetwork": NeuralNetwork,
     "RandomForest": RandomForest,
+    "RadialBasis": RadialBasis,
 }

autoemulate/emulators/radial_basis.py

@@ -38,43 +38,46 @@ def fit(self, X, y):
         self : object
             Returns self.
         """
-        X, y = check_X_y(X, y, multi_output=True, y_numeric=True)
+        X, y = check_X_y(X, y, multi_output=True, y_numeric=True, dtype=np.float64)
         self.n_features_in_ = X.shape[1]
-        self.model_ = RBF(d0=self.d0, poly_degree=self.poly_degree, reg=self.reg)
+        self.model_ = RBF(
+            d0=self.d0, poly_degree=self.poly_degree, reg=self.reg, print_global=False
+        )
         self.model_.set_training_values(X, y)
         self.model_.train()
+        self.is_fitted_ = True
         return self
 
-    def predict(self, X, return_std=False):
+    def predict(self, X):
         """Predicts the output of the emulator for a given input.
 
         Parameters
         ----------
         X : {array-like, sparse matrix}, shape (n_samples, n_features)
             The input samples.
 
-        return_std : bool, default=False
-            Whether to return the standard deviation of the prediction.
-
         Returns
         -------
         y : ndarray of shape (n_samples, n_features)
             The predicted values.
         """
-        X = check_array(X)
+        X = check_array(X, dtype=np.float64)
         check_is_fitted(self)
-        if return_std:
-            return self.model_.predict_values(X), self.model_.predict_variances(X)
-        else:
-            return self.model_.predict_values(X)
+        predictions = self.model_.predict_values(X)
+        # Check if dimension 2 is 1, if so, reshape to 1D array
+        # This is needed to pass sklearn's check_estimator
+        if predictions.ndim == 2 and predictions.shape[1] == 1:
+            predictions = predictions.ravel()
+        return predictions
 
     def get_grid_params(self):
         """Returns the grid parameters of the emulator."""
         param_grid = {
-            "d0": [1.0, 2.0, 3.0],
+            "d0": [1.0],
             "poly_degree": [-1, 0, 1],
             "reg": [1e-10, 1e-5, 1e-2],
         }
+        return param_grid
 
     def _more_tags(self):
         return {"multioutput": True}

tests/test_estimators.py

@@ -4,6 +4,7 @@
     GaussianProcessSk,
     NeuralNetwork,
     GaussianProcess,
+    RadialBasis,
 )
 from functools import partial
 
@@ -13,6 +14,7 @@
         RandomForest(random_state=42),
         GaussianProcessSk(random_state=1337),
         NeuralNetwork(random_state=13),
+        # RadialBasis(),
     ]
 )
 def test_check_estimator(estimator, check):