marcotcr / lime

Lime: Explaining the predictions of any machine learning classifier
BSD 2-Clause "Simplified" License
11.5k stars 1.79k forks source link

Is there way explain_instance can return multiple data points? #589

Closed lymphocyte closed 3 years ago

lymphocyte commented 3 years ago

Currently i use a for loop to get the explanations one by one like this:

exp = explainer.explain_instance(full_predict_data[i], predict_fn, num_features=5,num_samples=100)

And it's pretty slow to do so, Is there a way that I can get results of the whole datasets?

Thanks in advance.

marcotcr commented 3 years ago

Nope, that's the way to do it. sorry,