geek-yang
commented
9 months ago Hi Pim, thanks for your contribution. However, we modeled the .fit and .transform methods on sklearn, where these can be used in a pipeline. I think it would be better to do steps like cleaning up the data separately, instead of modifying the fit/transform methods.
An alternative way of adding this functionality, while not adding arguments to the .fit and .tranform methods, would be to add an argument to the
__init__method of the class.
Nearly forgot that when we designed this module, we would like to stay as close as possible to the sklearn pipeline recipe. Thanks for mentioning it!
Hi Pim, thanks for your contribution. However, we modeled the .fit and .transform methods on sklearn, where these can be used in a pipeline. I think it would be better to do steps like cleaning up the data separately, instead of modifying the fit/transform methods.
An alternative way of adding this functionality, while not adding arguments to the .fit and .tranform methods, would be to add an argument to the
__init__method of the class.