# Task 7. # Calculate TP, FP, TN, FN, Accuracy, Precision, Recall, and F-1 score # We assume that label y = 1 is positive, and y = 0 is negative def calc_metrics(Y_test, Y_pred_test): """ Calculate metrics Args: Y_test -- test label Y_pred_test -- predictions on test data Return: metrics -- a dict object """ assert(Y_test.shape == Y_pred_test.shape) ##### START YOUR CODE ##### TP = None FP = None TN = None FN = None Accuracy = None Precision = None Recall = None F1 = None ##### END YOUR CODE ##### metrics = { 'TP': TP, 'FP': FP, 'TN': TN, 'FN': FN, 'Accuracy': Accuracy, 'Precision': Precision, 'Recall': Recall, 'F1': F1 } return metrics

#### DO NOT CHANGE THE CODE BELOW #### # Evaluate Task 7 m = calc_metrics(Y_test, res['Y_pred_test']) print('TP = {}, FP = {}, TN = {}, FN = {}, Accuracy = {}, Precision = {}, Recall = {}, F1 = {}'.format( m['TP'], m['FP'], m['TN'], m['FN'], m['Accuracy'], m['Precision'], m['Recall'], m['F1'] ))

Calculate 8 evaluation metrics out of the previous results stored in the "res" object, using the ground truth label Ytest and the predictions on Ytest, which is stored in res['Y_pred_test']. NOTE: We assumte that label y=1 is positive, and y=0 is negative. \# Calculate TP, FP, TN, FN, Accuracy, Precision, Recall, and F-1 score \# We assume that Label y=1 is positive, and y=0 is negative Return: metrics - a dict object assert(Y_test.shape == Y_pred_test. shape) \#\#\#\# START YOUR CODE \#\#\#\#\# TP = None FP = None TN = None FN = None Accuracy = None Precision = None Recall = None F1 = None \#\#\#\# END YOUR CODE \#\#\#\#\# metrics = \{ 'TP': TP, 'FP': FP, 'TN': TN, 'FN': FN, 'Accuracy' : Accuracy, return metrics \#\#\# DO NOT CHANGE THE CODE BELOW \#\#\#\# \# Evaluate Task 7 m= calc_metrics(Y_test, res['Y_pred_test']) print ('TP ={},FP={},TN={},FN={},\ AAccuracy ={}, Precision ={},Recall={},F1={} format ( m[TP],m[ 'FP' ],m[TN],m[FN '], m[ 'Accuracy'], m[ 'Precision'], m[ 'Recall'], m[ 'F1'] ) ) Expected output TP=59FP=11TN=51FN=4Accuracy=0.88Precision=0.8428571428571429Recall=0.9365079365079365F1=0.887218045112782