Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Added the ability to graph distributions #11

Open
wants to merge 18 commits into
base: master
Choose a base branch
from
Open

Conversation

oobiwumi
Copy link
Collaborator

Added a function in audit_reading called graph_distributions that creates and saves distributions for all categorical vs numerical features, plotting the unrepaired distributions for each group along with the repaired distributions. In order to access the test split, I added lines in GradientFeatureAuditor.py that saves the test spit onto a file in the output directory so that graph_distributions can access them.

…raph,

but instead appears next to it.

Added the ability to run BBA auditor without writing to file while still
getting measure comparison graphs. Does not yet work with make_graphs.
Added option to print all data that would have been written to file
added labels to distribution graphs

cleaned up code
@oobiwumi
Copy link
Collaborator Author

Added option to not write to file while still printing all the data and graphs if the user wishes to do so.

@oobiwumi oobiwumi requested a review from sorelle June 27, 2019 20:15
yli11 added 14 commits June 28, 2019 14:02
numerical feature,  and optionally, the feature(s) you want graphed
… repaired

using graph_particular_distribution.
Mode, and Original.

Updated README

Fixed bug in graph_distributions
restricted values that can be put into repair_mode, defaulting to
"Orig" if invalid value is entered
Added repair_mode Spec, which allows the user to repair the data to
the group with the highest or lowest median

Updated README
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
2 participants