'modify'

smasky · May 7, 2024 · 85ec965 · 85ec965
1 parent a2e6314
commit 85ec965
Show file tree

Hide file tree

Showing 3 changed files with 26 additions and 25 deletions.
diff --git a/UQPyL/problems/problem_ABC.py b/UQPyL/problems/problem_ABC.py
@@ -48,6 +48,7 @@ def _check_2d(self, X:np.ndarray):
 
     def _check_bound(self,bound: np.ndarray):
 
-        if(not bound.shape[1]==self.n_input):
+        bound_tmp=bound.ravel()
+        if(not bound.shape[0]==self.n_input):
             raise ValueError('the input bound is inconsistent with the input n_inputensions')
 
diff --git a/UQPyL/surrogates/gp_kernels/base_kernel.py b/UQPyL/surrogates/gp_kernels/base_kernel.py
@@ -3,9 +3,9 @@
 
 class Gp_Kernel():
     def __init__(self):
-        self._theta=None
-        self._theta_ub=None
-        self._theta_lb=None
+        self._theta={}
+        self._theta_ub={}
+        self._theta_lb={}
 
     def __check_array__(self, value: Union[float,np.ndarray]):
 

diff --git a/examples/example_surrogates.py b/examples/example_surrogates.py
@@ -38,27 +38,27 @@
 #test set
 X_test=lhs.sample(50, 10)
 Y_test=problem.evaluate(X_test, unit=True)
-###########################################
-
-#############1. full_connect_neural_network (FNN)#################
-print("#################full_connect_neural_network (FNN)#################")
-from UQPyL.surrogates import FNN
-#use 0-1 MinMaxScaler to normalize data
-fnn=FNN(scalers=(MinMaxScaler(0,1), MinMaxScaler(0,1)),
-        hidden_layer_sizes=[16,32,64,32,16,8], 
-        activation_functions='relu', solver='adam', alpha=0.001)
-fnn.fit(X_train,Y_train)
-Y_predict=fnn.predict(X_test)
-#use R-square to validate the Y_predict and Y_test
-
-r2=r2_score(Y_test, Y_predict)
-print('r2:', r2)
-#use rank_score to validate the Y_predict and Y_test
-
-oe=rank_score(Y_test, Y_predict)
-print('rank_score:', oe)
-
-############2. gaussian_process (GP)#################
+# ###########################################
+
+# #############1. full_connect_neural_network (FNN)#################
+# print("#################full_connect_neural_network (FNN)#################")
+# from UQPyL.surrogates import FNN
+# #use 0-1 MinMaxScaler to normalize data
+# fnn=FNN(scalers=(MinMaxScaler(0,1), MinMaxScaler(0,1)),
+#         hidden_layer_sizes=[16,32,64,32,16,8], 
+#         activation_functions='relu', solver='adam', alpha=0.001)
+# fnn.fit(X_train,Y_train)
+# Y_predict=fnn.predict(X_test)
+# #use R-square to validate the Y_predict and Y_test
+
+# r2=r2_score(Y_test, Y_predict)
+# print('r2:', r2)
+# #use rank_score to validate the Y_predict and Y_test
+
+# oe=rank_score(Y_test, Y_predict)
+# print('rank_score:', oe)
+
+# ############2. gaussian_process (GP)#################
 print("#################gaussian_process (GP)#################")
 from UQPyL.surrogates.gaussian_process import GPR
 # we should construct a kernel for GPR