-1

I have a a confusion matrix and a classification report as shown in the attached image. My question is, how this accuracy is calculated? I have tried many formula to calculate the accuracy by hand but no one give me the right one?

please, I need a help to find how the model is calculate this accuracy? What is the mathematical equation that he used to reach this accuracy? enter image description here

Timbus Calin
  • 13,809
  • 5
  • 41
  • 59
ALI
  • 1
  • 2

1 Answers1

0

You need to divide the total number of correct answers by your total sample size. In this case:

  • class 0: 918 * 0.9434... = 866 correct predictions
  • class 1: 2522 * 0.9964... = 2513
  • class 2: 349 * 0.9198... = 321
  • (866 + 2513 + 321) / 3789 = 0.9765...
dx2-66
  • 2,376
  • 2
  • 4
  • 14