pmlearn/gaussian_process/gpr.py

Survived

Mutant 10

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -48,7 +48,7 @@
             f_pred = self.gp.conditional("f_pred", X)
             self.ppc = pm.sample_ppc(self.trace,
                                      vars=[f_pred],
-                                     samples=2000)
+                                     samples=2001)
 
         if return_std:
             return self.ppc['f_pred'].mean(axis=0), \

Mutant 18

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -322,7 +322,7 @@
     """
 
     def __init__(self, prior_mean=None, kernel=None):
-        self.ppc = None
+        self.ppc = ""
         self.gp = None
         self.num_training_samples = None
         self.num_pred = None

Mutant 20

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -324,7 +324,7 @@
     def __init__(self, prior_mean=None, kernel=None):
         self.ppc = None
         self.gp = None
-        self.num_training_samples = None
+        self.num_training_samples = ""
         self.num_pred = None
         self.prior_mean = prior_mean
         self.kernel = kernel

Mutant 22

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -326,7 +326,7 @@
         self.gp = None
         self.num_training_samples = None
         self.num_pred = None
-        self.prior_mean = prior_mean
+        self.prior_mean = None
         self.kernel = kernel
 
         super(SparseGaussianProcessRegressor, self).__init__()

Mutant 23

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -327,7 +327,7 @@
         self.num_training_samples = None
         self.num_pred = None
         self.prior_mean = prior_mean
-        self.kernel = kernel
+        self.kernel = None
 
         super(SparseGaussianProcessRegressor, self).__init__()
 

Mutant 28

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -352,7 +352,7 @@
             'model_output': model_output,
         }
 
-        self.gp = None
+        self.gp = ""
         model = pm.Model()
 
         with model:

Mutant 30

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -356,7 +356,7 @@
         model = pm.Model()
 
         with model:
-            length_scale = pm.Gamma('length_scale', alpha=2, beta=1,
+            length_scale = pm.Gamma('XXlength_scaleXX', alpha=2, beta=1,
                                     shape=(1, self.num_pred))
             signal_variance = pm.HalfCauchy('signal_variance', beta=5,
                                             shape=1)

Mutant 31

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -356,7 +356,7 @@
         model = pm.Model()
 
         with model:
-            length_scale = pm.Gamma('length_scale', alpha=2, beta=1,
+            length_scale = pm.Gamma('length_scale', alpha=3, beta=1,
                                     shape=(1, self.num_pred))
             signal_variance = pm.HalfCauchy('signal_variance', beta=5,
                                             shape=1)

Mutant 32

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -356,7 +356,7 @@
         model = pm.Model()
 
         with model:
-            length_scale = pm.Gamma('length_scale', alpha=2, beta=1,
+            length_scale = pm.Gamma('length_scale', alpha=2, beta=2,
                                     shape=(1, self.num_pred))
             signal_variance = pm.HalfCauchy('signal_variance', beta=5,
                                             shape=1)

Mutant 35

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -358,7 +358,7 @@
         with model:
             length_scale = pm.Gamma('length_scale', alpha=2, beta=1,
                                     shape=(1, self.num_pred))
-            signal_variance = pm.HalfCauchy('signal_variance', beta=5,
+            signal_variance = pm.HalfCauchy('XXsignal_varianceXX', beta=5,
                                             shape=1)
             noise_variance = pm.HalfCauchy('noise_variance', beta=5,
                                            shape=1)

Mutant 40

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -360,7 +360,7 @@
                                     shape=(1, self.num_pred))
             signal_variance = pm.HalfCauchy('signal_variance', beta=5,
                                             shape=1)
-            noise_variance = pm.HalfCauchy('noise_variance', beta=5,
+            noise_variance = pm.HalfCauchy('noise_variance', beta=6,
                                            shape=1)
 
             if self.kernel is None:

Mutant 44

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -364,7 +364,7 @@
                                            shape=1)
 
             if self.kernel is None:
-                cov_function = signal_variance ** 2 * RBF(
+                cov_function = signal_variance * 2 * RBF(
                     input_dim=self.num_pred,
                     ls=length_scale)
             else:

Mutant 45

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -364,7 +364,7 @@
                                            shape=1)
 
             if self.kernel is None:
-                cov_function = signal_variance ** 2 * RBF(
+                cov_function = signal_variance ** 3 * RBF(
                     input_dim=self.num_pred,
                     ls=length_scale)
             else:

Mutant 52

--- pmlearn/gaussian_process/gpr.py
+++ pmlearn/gaussian_process/gpr.py
@@ -381,7 +381,7 @@
 
             # initialize 20 inducing points with K-means
             # gp.util
-            Xu = pm.gp.util.kmeans_inducing_points(20,
+            Xu = pm.gp.util.kmeans_inducing_points(21,
                                                    X=model_input.get_value())
 
             y = self.gp.marginal_likelihood('y',