Python Programming

Let’s do one more task!

Here is our confusion matrix again.

For a Machine Learning classification task, we might have to compute an evaluation metric called $accuracy$.

$accuracy = \frac{| correct |}{| instances |}$

Your task

Use NumPy to compute the accuracy from the confusion matrix above.

As hinted in the equation, you will need the compute:

1. the total number of correct predictions
2. the total number of instances
3. divide (1) by (2)

Implement this in NumPy. It can be done in a single line if you wish!

import numpy as np

x = np.array([[18, 3, 9], [1, 30, 4], [7, 5, 23]])
accuracy = ????
assert accuracy == 0.71

👀 Peek at sample solutions

accuracy = x.diagonal().sum() / x.sum()
accuracy = np.sum(np.diagonal(x)) / np.sum(x)