preprocessing.cleaning
logger_config
logger
lowconf_cleaning
def lowconf_cleaning(config: dict,
read_from_variable: str = "position_processed",
save_to_variable: str = "position_processed") -> None
Clean the low confidence data points from the dataset. Processes position data by:
- setting low-confidence points to NaN
- interpolating NaN points
Parameters
- config (
dict
): Configuration dictionary. - read_from_variable (
str, optional
): Variable to read from the dataset. - save_to_variable (
str, optional
): Variable to save the cleaned data to.
Returns
None
outlier_cleaning
def outlier_cleaning(config: dict,
read_from_variable: str = "position_processed",
save_to_variable: str = "position_processed") -> None
Clean the outliers from the dataset. Processes position data by:
- setting outlier points to NaN
- interpolating NaN points
Parameters
- config (
dict
): Configuration dictionary. - read_from_variable (
str, optional
): Variable to read from the dataset. - save_to_variable (
str, optional
): Variable to save the cleaned data to.
Returns
None