Sklearn Precision and recall giving wrong values

Question

Why is my precision score so low in the above image?

Answer 1

I see in your comments that you're trying to interpret confusion_matrix as [[tp, fp], [fn, tn]]

Based on documentation, sklearn.confusion_matrix is a function that returns an array of:

[[tn, fp], [fn, tp]]

So, it's vice-versa and the calculation is right:

397 / (397 + 925) = 0.30030257...