mydatapreprocessing.preprocessing.preprocessing_config package¶
Preprocessing config and subconfig classes.
-
mydatapreprocessing.preprocessing.preprocessing_config.
default_preprocessing_config
¶ Default config, that you can use. You can use intellisense with help tooltip to see what you can setup there or you can use update method for bulk configuration.
Type: mydatapreprocessing.preprocessing.preprocessing_config.PreprocessingConfig
-
class
mydatapreprocessing.preprocessing.preprocessing_config.
PreprocessingConfig
(frozen=None, *a, **kw)[source]¶ Bases:
mypythontools.config.config_internal.Config
Config class for preprocess_data pipeline.
There is default_preprocessing_config object already created. You can import it, edit and use. Static type check and intellisense should work.
-
difference_transform
¶ Transform the data.
Type
bool
Default
False
‘difference’ transform data into differences between neighbor values.
-
remove_outliers
¶ Remove unusual values far from average.
Type
None | Numeric
Default
None
-
smooth
¶ Smooth the data with Savitzky-Golay filter.
Type
None | tuple[int, int]
Default
None
Setup with tuple (window, polynomial_order) as in smooth function e.g (11, 2).
-
standardize
¶ Standardize the data.
Type
None | Literal[“standardize”, “-11”, “01”, “robust”]
Default
‘standardize’
‘01’ and ‘-11’ means scope from to for normalization. ‘robust’ use RobustScaler and ‘standard’ use StandardScaler - mean is 0 and std is 1. If no standardization, use None.
-