基于两个边界框列表计算精度和召回率

Question

does anyone know, given two lists of the bounding box, one is predicted, the other is the ground truth, is there some existing python library or code I can use to compute the accuracy? There are many existing codes for first calculating the lou, and I know I can compute it on my own. But I thought there should also be some existing tools I can use to directly get the accuracy indexes so that I finish my project in a smarter way than coding by myself.

Thanks!

Answer 1

Yes there is a lot of love from sklearn.metrics

Heres a link: http://scikit-learn.org/stable/modules/classes.html#sklearn-metrics-metrics

Here's a basic example:

from sklearn.metrics import accuracy_score
pr_score = accuracy_score(y_test, y_pred)
print(pr_score)

Where y_pred is the predictions and y_test is the so called "ground truth"

In the above link you can read and chose the evaluation metric that suits you.