You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As a user, it would be useful to have a constraint that specifies the lowest common increment the data can appear in. For example, if my data must be a multiple of 5 or 7, it would be useful to be able to specify this.
Expected behavior
Remove the Rounding constraint. This functionality should be handled by the HyperTransformer
Add a FixedIncrements constraint
Init params should be:
column_name: String that is the name of the column to be constrained
increment_value: Int that is the value to make sure all data should be divisible by. Must be greater than 0.
Validation
If the increment_value is less than 0 or 0, we should raise an error during initialization.
If the data is not evenly divisible by the increment_value then that row is invalid
Nans should be ignored during validation
reverse_transform
Round synthesized values to nearest multiple of increment_value
Additional context
one option for transform is to divide all values by increment_value. Then on reverse you can round to the nearest int and just multiply.
The text was updated successfully, but these errors were encountered:
Problem Description
As a user, it would be useful to have a constraint that specifies the lowest common increment the data can appear in. For example, if my data must be a multiple of 5 or 7, it would be useful to be able to specify this.
Expected behavior
Rounding
constraint. This functionality should be handled by theHyperTransformer
FixedIncrements
constraintcolumn_name
: String that is the name of the column to be constrainedincrement_value
: Int that is the value to make sure all data should be divisible by. Must be greater than 0.increment_value
is less than 0 or 0, we should raise an error during initialization.increment_value
then that row is invalidreverse_transform
increment_value
Additional context
increment_value
. Then on reverse you can round to the nearest int and just multiply.The text was updated successfully, but these errors were encountered: