Avoid Error When There Is No Positive Class In Certain Bins for KL Divergence Calculation

[lleiou]

Proposed changes

For KL Divergence, we will calculate the probability of conversion in each bin first in _GetNodeSummary(), in order to do that we will need to know the number of positive population under each treatment or control group, here is how the code currently gets the number:

n_1 = results[treatment_group_key][1]

However, there are rare cases when there is no positive population within a bin for a given treatment or control group and the code above will break, so I propose to change it to the following so that when there is no positive population then n_1 will be 0:

n_1 = results[treatment_group_key].get(1, 0)

Types of changes

What types of changes does your code introduce to CausalML?
Put an x in the boxes that apply

• Bugfix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation Update (if none of the other choices apply)

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[paullo0106]

LGTM, I was able to reproduce the issue when # of bins is large and the data size is small, and verified the fix.

Thanks for the contribution!

[lleiou]

Thank you!