Regression Example about lime HOT 5 CLOSED

Hi @joshualeond & @thomasp85,

I am having the exact same problem as Joshua with my random forest.
I don't know what to put as "labels" and "n_labels" in order to solve it?
Did you manage?

Thank you very much in advance.
Maroussia

from lime.

You seem to have stumbled upon a bug - I'll look into it. Thanks

from lime.

Hey @Marouds, I was able to run this example without any errors. This original bug resulted in the error:
#> Error in y[1, ]: incorrect number of dimensions

You're receiving this same error? Are you receiving this error using the example above or are you working with different data?

For a regression model you can remove the labels and n_labels arguments from the function call. Either way though, the explain() function should run without errors on this example.

from lime.

Hi @joshualeond, thank you very much for your quick reply!

I managed to run it in the end !! And indeed I was receiving the same error "#> Error in y[1, ]: incorrect number of dimensions". I added "drop=FALSE" when I split my data into training and testing dataset, and it worked!
(I might be using an old version of R)

Thanks!
Have a nice day

from lime.

I do have an issue with the visualization of the results, using plot_features().
It does not appear with colors and everything like it seems it should...
I don't know whether an addition packages was needed?
Thanks!!

from lime.