test

total_combinations = generateCombinations(features,number_per_combination)

for feature_combination in total_combinations:
    
    X = data[list(feature_combination)]
    sc_X = StandardScaler()
    df = sc_X.fit_transform(X)
    clf =LogisticRegression()
    clf.fit(df,y)
    train_score = clf.score(df,y)
    
    if(train_score.mean()>0.8):
        
        X = data[list(feature_combination)]
        sc_X = StandardScaler()
        df = sc_X.fit_transform(X)
        clf =LogisticRegression()
        cross_validation_score= cross_val_score(clf, X, y, cv=7)
        
        if(cross_validation_score.mean()>0.7):
            
            finalfeatureholder.append(str(feature_combination))
            scoreholder.append(str(train_score.mean()))
            CVMRscoreholder.append(str(cross_validation_score.mean()))
            CVMRstdholder.append(str(cross_validation_score.std()*2))