-
Notifications
You must be signed in to change notification settings - Fork 923
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
Create estimation validation loop #136
Create estimation validation loop #136
Conversation
We create an estimator for t=t_i and then find the value for the other values of t_i. This allows us to isolate the causal effect of t from y during the optimization process.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good to me so far. Added a suggestion on how to simply data_preprocess method.
I guess you are now working on running the dummy outcome generator on the other chunks and then evaluating the effect?
In a nutshell, the plan is to train on a partition of a |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The logic is correct, but can be simplified in many cases.
Also there is a larger point of what to do when the pipeline actions contain only zero, noise or permute. does not make sense to create groups or validation set. I gave a suggestion in the inline comments. Let me know what you think.
- move func_args to the end of argument list - replace X_chunk by X_train
All function calls now pass func_args as a variable length keyworded argument list ( **kwargs ). Accordingly, all function calls and examples in the documentation have been changed.
- Move the no_estimator check out of the simulation loop - Replace np.zeros with None - Use a common variable validation_df to refer to the validation data
Thanks @Tanmay-Kulkarni101 . This looks ready to merge once you add a few tests. |
Change the behavior of Dummy Outcome Refuter, so as to compare the behavior, keeping the value of the treatment constant.