casact
[ENH] Adding Pipeline support to Development
Is your feature request related to a problem? Please describe. within the default development class, ata selections (drop, n_period, etc.) is performed via a fixed order. custom order is not possible since development currently does not use the previous ata selections (e.g. Development(n_period = 4).fit_transform(Development(drop_low = 1).fit_transform(tri)) is not equal to Development(n_period = 4,drop_low = 1).fit_transform(tri))
Is your feature request at odds with the scope of the package? this feature is strictly for reserving and would offer greater alignment with the sklearn API
Describe the solution you'd like let the previous ata selections presist when further transforming
Describe alternatives you've considered parameterize order of selection to eliminate the need of chaining/pipelining development transformers. messy
Additional context probably clean up ata selection logic in the meantime
