GCM.py

#####
# Do multi-classification task,
# then get the ground truth and the predict label named y_true and y_pred
from sklearn import metrics
import numpy as np

y_pred = [1,2,3,4,5,1,2,1,1,4,5]
y_true = [1,2,3,4,5,1,2,1,1,4,1]
def kappa(confusion_matrix, k):
    dataMat = np.mat(confusion_matrix)
    P0 = 0.0
    for i in range(k):
        P0 += dataMat[i, i]*1.0
    xsum = np.sum(dataMat, axis=1)
    ysum = np.sum(dataMat, axis=0)
    Pe  = float(ysum*xsum)/np.sum(dataMat)**2
    OA = float(P0/np.sum(dataMat)*1.0)
    cohens_coefficient = float((OA-Pe)/(1-Pe))
    return cohens_coefficient

classify_report = metrics.classification_report(y_true, y_pred)
confusion_matrix = metrics.confusion_matrix(y_true, y_pred)
overall_accuracy = metrics.accuracy_score(y_true, y_pred)
acc_for_each_class = metrics.precision_score(y_true, y_pred, average=None)
average_accuracy = np.mean(acc_for_each_class)
kappa_coefficient = kappa(confusion_matrix, 5)
print('classify_report : \n', classify_report)
print('confusion_matrix : \n', confusion_matrix)
print('acc_for_each_class : \n', acc_for_each_class)
print('average_accuracy: {0:f}'.format(average_accuracy))
print('overall_accuracy: {0:f}'.format(overall_accuracy))
print('kappa coefficient: {0:f}'.format(kappa_coefficient))