Refactoring

Raw Code/.gitignore

+../idea/
+../multiview-machine-learning-omis.iml
\ No newline at end of file

Raw Code/FeatExtraction/ClassifMonoView.py

@@ -3,6 +3,7 @@
 """ MultiClass Classification with MonoView """
 
 # Import built-in modules
+import pandas as pd
 
 # Import sci-kit learn party modules
 from sklearn.cross_validation import train_test_split

Raw Code/FeatExtraction/ExecFeatExtraction.py

@@ -126,7 +126,6 @@ CELL_DIMENSION = 5
 NB_ORIENTATIONS = 8
 NB_CLUSTERS = 12
 MAXITER = 100
-NB_CORES = 1
 
 # Extract Feature from DB
 hog_feat_desc,hog_f_extr_res = FeatExtraction.calcHOGParallel(nameDB, dfImages.values, CELL_DIMENSION, NB_ORIENTATIONS, NB_CLUSTERS, MAXITER, NB_CORES)

Raw Code/Fusion/EarlyFusion.py

@@ -3,6 +3,7 @@
 
 import numpy as np
 
+
 # TODO :
 # Linear Weighted Fusion
 # Bayesian Inference  /!\ Statistically independant => ? 
@@ -15,13 +16,14 @@ import numpy as np
 # Particle Filter ?
 
 def linearWeightedFusion(toFuse, weights):
-	
     # Normalize weights ?
     # weights = weights/float(max(weights))
 
-	weighted = np.array([np.array([feature*weights for (feature, weight) in zip(exampleToFuse, weights)]).flatten() for exampleToFuse in toFuse])
-	return fused
+    weighted = np.array(
+            [np.array([feature * weights for (feature, weight) in zip(exampleToFuse, weights)]).flatten() for
+             exampleToFuse in toFuse])
+    return weighted
 
 
 if __name__ == '__main__':
-	
\ No newline at end of file
+    pass

Raw Code/Fusion/LateFusion.py

@@ -5,6 +5,7 @@ import numpy as np
 import sys
 from sklearn.svm import SVC
 
+
 # Our method in multiclass classification will be One-vs-One or One-vs-All
 # classifiers, so if we can get the output of these classifiers, we are 
 # able to compute a score for each class in each mono-view classification
@@ -17,18 +18,17 @@ from sklearn.svm import SVC
 
 # weights : (nbFeature) array with the weights for each feature
 def weightedLinear(monoViewDecisions, weights):
-
     # Normalize weights ?
     # weights = weights/float(max(weights))
 
     fusedExamples = np.array([sum(np.array([featureScores * weight for weight, featureScores \
-					 	in zip(weights, exampleDecisions)])) for exampleDecisions in monoViewDecisions])
+                                            in zip(weights, exampleDecisions)])) for exampleDecisions in
+                              monoViewDecisions])
     # print fused
 
     return np.array([np.argmax(fusedExample) for fusedExample in fusedExamples])
 
 
-
 # The SVMClassifier is here used to find the right weights for linear fusion
 def SVMForLinearFusionTrain(monoViewDecisions, labels):
     SVMClassifier = SVC()
@@ -41,7 +41,6 @@ def SVMForLinearFusionFuse(monoViewDecisions, SVMClassifier):
     return labels
 
 
-
 # For majority voting, we have a problem : we have 5 fetures and 101 classes
 # on Calthech, so if each feature votes for one class, we can't find a good 
 # result
@@ -66,8 +65,6 @@ def majorityVoting(monoViewDecisions, NB_CLASS):
     return np.array([np.argmax(exampleVotes) for exampleVotes in votes])
 
 
-
-
 # Main for testing
 if __name__ == '__main__':
     DATASET_LENGTH = 10
@@ -77,20 +74,23 @@ if __name__ == '__main__':
     LABELS = np.array([TRUE_CLASS for i in range(DATASET_LENGTH)])
     LABELS[0] = 0
 
-	monoViewDecisionsEasy = np.array([np.array([np.zeros(NB_CLASS) for i in range(nbFeature)])for example in range(DATASET_LENGTH)])
+    monoViewDecisionsEasy = np.array(
+            [np.array([np.zeros(NB_CLASS) for i in range(nbFeature)]) for example in range(DATASET_LENGTH)])
     for exampleDecisions in monoViewDecisionsEasy:
         for decision in exampleDecisions:
             decision[TRUE_CLASS] = 12
     # print monoViewDecisionsEasy
 
-	monoViewDecisionsHard = np.array([np.array([np.zeros(NB_CLASS) for i in range(nbFeature)])for example in range(DATASET_LENGTH)])
+    monoViewDecisionsHard = np.array(
+            [np.array([np.zeros(NB_CLASS) for i in range(nbFeature)]) for example in range(DATASET_LENGTH)])
     for exampleDecisions in monoViewDecisionsHard:
         for decision in exampleDecisions:
             decision[TRUE_CLASS] = 12
         exampleDecisions[nbFeature - 2] = np.zeros(NB_CLASS) + 1400
         exampleDecisions[nbFeature - 2][TRUE_CLASS] -= 110
 
-	monoViewDecisionsMajority = np.array([np.array([TRUE_CLASS,TRUE_CLASS,TRUE_CLASS,1,5]) for example in range(DATASET_LENGTH)])
+    monoViewDecisionsMajority = np.array(
+            [np.array([TRUE_CLASS, TRUE_CLASS, TRUE_CLASS, 1, 5]) for example in range(DATASET_LENGTH)])
     monoViewDecisionsMajorityFail = np.array([np.array([1, 2, 3, 4, 5]) for example in range(DATASET_LENGTH)])
 
     weights = np.random.rand(nbFeature)
@@ -98,7 +98,6 @@ if __name__ == '__main__':
 
     SVMClassifier = SVMForLinearFusionTrain(monoViewDecisionsMajority, LABELS)
 
-
     print weightedLinear(monoViewDecisionsEasy, weights)
     print weightedLinear(monoViewDecisionsHard, weights)
     print SVMForLinearFusionFuse(monoViewDecisionsMajority, SVMClassifier)