The training data contain 10 positive examples (y = 1) and 10 negative examples (y = 0).

8 out of the 10 positive examples are correctly predicted, and 6 out of the 10 negative examples are correctly predicted.

Fill in the following blanks about the metrics:

below is wrong.

True Positive(TP) - 8

False Positive(FP) - 2

True negative(TN) - 6

False negative(FN) - 4

Accuracy - (TP + TN ) / ( TP + FP + TN + FN )

Hence, Accuracy = 14 / 20 = 0.7

Accuracy = 0.7

Precision - TP / TP + FP

Hence, precision = 8 / 10 = 0.80

Precision = 0.80

Recall - TP / TP+ FN

hence, recall = 8 / 12 = 0.67

Recall = 0.67

F1 Score - 2*((precision * recall) / ( precision + recall))

Hence, F1 Score = 0.73

below is also wrong.

True Positive Count (TP) =8 False Positive Count (FP) =4 True Negative Count (TN)=6 False Negative Count (FN): 2 Accuracy=TP+TN+FP+FNTP+TN=2014=70%Answer Precision=TP+FPTP=128=66.66%Answer Recall=TP+FNTP=108=80%Answer F1score=2Precision+RecallPrecisionRecall 2128+108128108=36.30%