def DamagePrediction(mutation): FeatVec = [] for label in D_Labels: FeatVec.append(mutation[label]) D_Tree = pickle.load(open(D_TreeFile)) Pred = DecisionTree.Classify(D_Tree, D_Labels, FeatVec) return Pred
def ConservePrediction(mutation): FeatVec = [] for label in C_Labels: FeatVec.append(mutation[label]) C_Tree = pickle.load(open(C_TreeFile)) Pred = DecisionTree.Classify(C_Tree, C_Labels, FeatVec) return Pred
C_Positive_t = GetPositiveFeatureVector('testing.xls',C_Labels,4) D_Positive_t = GetPositiveFeatureVector('testing.xls',D_Labels,4) C_Negative,C_TestVec = GetNegativeFeatureVector('10128.annotation.xls',C_Labels) D_Negative,D_TestVec = GetNegativeFeatureVector('10128.annotation.xls',D_Labels) ############################################################################# print len(D_Positive) C_dataset = C_Negative + C_Positive D_dataset = D_Negative + D_Positive #print dataset C_Tree = DecisionTree.CreatTree(C_dataset,C_Labels) D_Tree = DecisionTree.CreatTree(D_dataset,D_Labels) #TreePlotter.CreatPlot(D_Tree) #TreePlotter.CreatPlot(C_Tree) C_TreeFile = open('C_Tree','w') D_TreeFile = open('D_Tree','w') pickle.dump(C_Tree,C_TreeFile) pickle.dump(D_Tree,D_TreeFile) C_TreeFile.close();D_TreeFile.close() ClassList = [] #print Positive[50:] # [50:]: #Positive: #TestVec: for vec in C_TestVec: vec = vec[0:3] ClassList.append(DecisionTree.Classify(C_Tree,C_FeatLabels,vec)) print len(ClassList),ClassList.count('no')/len(ClassList)