Repository navigation
plot_shap
#247
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
I have a very short question:
Why does plot_shap call "shap_values = shap_values[:, :, 0]", doesnt this lead to the resulting plots (e.g., beeswarm plots) giving results for the "negative" class (assuming 0 is the negative class in the data)? For example in my dataset, 0 = no remission and 1 = remission from a mental illness. Using plot_shap would yield a beeswarm plot where positive values for a sample-feature combination would indicate higher probability for non remission correct? I see that this is not wrong but i wonder why this approach was chosen as it may confuse people (at least i was confused).
All reactions