MKL adaptation ongoing

config_files/config_test.yml

 # The base configuration of the benchmark
 Base :
-  log: true
-  name: ["Plausible"]
+  log: True
+  name: ["control_vs_malade"]
   label: "_"
   type: ".hdf5"
   views:
@@ -18,18 +18,18 @@ Base :
 # All the classification-realted configuration options
 Classification:
   multiclass_method: "oneVersusOne"
-  split: 0.8
-  nb_folds: 2
+  split: 0.4
+  nb_folds: 5
   nb_class: 2
   classes:
   type: ["multiview"]
   algos_monoview: ["all"]
   algos_multiview: ["lp_norm_mkl",]
-  stats_iter: 2
+  stats_iter: 5
   metrics: ["accuracy_score", "f1_score"]
   metric_princ: "f1_score"
   hps_type: "randomized_search"
-  hps_iter: 2
+  hps_iter: 30
 
 
 #####################################

multiview_platform/mono_multi_view_classifiers/multiview_classifiers/additions/kernel_learning.py

@@ -68,7 +68,7 @@ class KernelConfigDistribution:
     def __init__(self, seed=42):
         self.random_state=np.random.RandomState(seed)
         self.possible_config = {
-            "additive_chi2": {"gamma": CustomUniform()},
+            "chi2": {"gamma": CustomUniform()},
             "rbf": {"gamma": CustomUniform()},
         }
 
@@ -95,7 +95,7 @@ class KernelDistribution:
 
     def __init__(self, seed=42):
         self.random_state=np.random.RandomState(seed)
-        self.available_kernels = ["rbf",]
+        self.available_kernels = ["rbf"]
 
     def draw(self, nb_view):
         return list(self.random_state.choice(self.available_kernels, nb_view))

multiview_platform/mono_multi_view_classifiers/multiview_classifiers/lp_norm_mkl.py

@@ -31,7 +31,8 @@ class LPNormMKL(KernelClassifier, MKL):
         return super(LPNormMKL, self).fit(formatted_X, y[train_indices])
 
     def predict(self, X, example_indices=None, view_indices=None):
-        print(self.C.shape)
+        # print(self.K_)
+        # import pdb;pdb.set_trace()
         new_X, _ = self.format_X(X, example_indices, view_indices)
         return self.extract_labels(super(LPNormMKL, self).predict(new_X))