dropna #178
dropna #178
Conversation
| @@ -209,13 +209,16 @@ def __init__( # noqa: PLR0913 | |||
| self._trend: dict | |||
| self._is_fit = False | |||
|
|
|||
| def fit(self, data: Union[xr.DataArray, xr.Dataset]) -> None: | |||
| def fit(self, data: Union[xr.DataArray, xr.Dataset], dropna=False) -> None: | |||
There was a problem hiding this comment.
| def fit(self, data: Union[xr.DataArray, xr.Dataset], dropna=False) -> None: | |
| def fit(self, data: Union[xr.DataArray, xr.Dataset], dropna: bool = False) -> None: |
Also add a type here to please mypy.
| @@ -241,16 +243,20 @@ def fit(self, data: Union[xr.DataArray, xr.Dataset]) -> None: | |||
| self._is_fit = True | |||
|
|
|||
| def transform( | |||
| self, data: Union[xr.DataArray, xr.Dataset] | |||
| self, data: Union[xr.DataArray, xr.Dataset], dropna=False | |||
There was a problem hiding this comment.
| self, data: Union[xr.DataArray, xr.Dataset], dropna=False | |
| self, data: Union[xr.DataArray, xr.Dataset], dropna: bool = False |
| """Fit this Preprocessor to input data. | ||
|
|
||
| Args: | ||
| data: Input data for fitting. | ||
| dropna: If True, drop all NaN values from the data before preprocessing. |
There was a problem hiding this comment.
| dropna: If True, drop all NaN values from the data before preprocessing. | |
| dropna: If True, drop all NaN values from the data along time axis before preprocessing. |
Point out that this dropna works only for the time axis.
| ) -> Union[xr.DataArray, xr.Dataset]: | ||
| """Apply the preprocessing steps to the input data. | ||
|
|
||
| Args: | ||
| data: Input data to perform preprocessing. | ||
| dropna: If True, drop all NaN values from the data before preprocessing. |
There was a problem hiding this comment.
| dropna: If True, drop all NaN values from the data before preprocessing. | |
| dropna: If True, drop all NaN values from the data along time axis before preprocessing. |
|
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 |
Nearly forgot that when we designed this module, we would like to stay as close as possible to the |
No description provided.