1
$\begingroup$

I'm using LIME to explain my random forest model. Everything is working great. However, I don't quite understand the image that is generated. Taking the example from the Readme:

enter image description here

How can it predict poisonous with 1.00 but still have a gill-size=broad which has 0.13 probability? Or do I misunderstand the table?

I already asked this question in the github project, but didn't get any answer.

$\endgroup$
1
$\begingroup$

I think the answer mostly lies in the fact that these are just approximations and they're not super exact because of the small data set and nature of decision trees. The prediction was really 1.0 (I'm guessing all trees' leaves agreed entirely on the prediction).

If gill_size != broad, it would still be 1.0, you could say (1.13 is meaningless). But maybe if gill_size != broad and odor != foul you could expect the probability to be about 1.0 + 0.13 - 0.26 = 0.87.

A little more explanation is at https://marcotcr.github.io/lime/tutorials/Tutorial%20-%20continuous%20and%20categorical%20features.html

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.