s2spy.preprocess

Preprocessor for s2spy workflow.

Module Contents

Classes

Preprocessor

Preprocessor for s2s data.

Functions

_linregress(→ tuple[float, float])	Calculate the slope and intercept between two arrays using scipy's linregress.
_trend_linear(→ dict)	Calculate the linear trend over time.
_subtract_linear_trend(data, trend)	Subtract a previously calclulated linear trend from (new) data.
_get_trend(data, method)	Calculate the trend, with a certain method. Only linear is implemented.
_subtract_trend(data, method, trend)	Subtract the previously calculated trend from (new) data. Only linear is implemented.
_get_climatology(data, timescale)	Calculate the climatology of timeseries data.
_subtract_climatology(data, timescale, climatology)
_check_input_data(data)	Check the input data for compatiblity with the preprocessor.
_check_temporal_resolution(→ Literal[monthly, weekly, ...)
_check_data_resolution_match(data, timescale)	Check if the temporal resolution of input is the same as given timescale.

s2spy.preprocess._linregress(x: numpy.ndarray, y: numpy.ndarray) → tuple[float, float]

Calculate the slope and intercept between two arrays using scipy’s linregress.

Used to make linregress more ufunc-friendly.

Parameters:

• x – First array.

• y – Second array.

Returns:

slope, intercept

s2spy.preprocess._trend_linear(data: xarray.DataArray | xarray.Dataset) → dict

Calculate the linear trend over time.

Parameters:

data – The input data of which you want to know the trend.

Returns:

Dictionary containing the linear trend information (slope and intercept)

s2spy.preprocess._subtract_linear_trend(data: xarray.DataArray | xarray.Dataset, trend: dict)

Subtract a previously calclulated linear trend from (new) data.

s2spy.preprocess._get_trend(data: xarray.DataArray | xarray.Dataset, method: str)

Calculate the trend, with a certain method. Only linear is implemented.

s2spy.preprocess._subtract_trend(data: xarray.DataArray | xarray.Dataset, method: str, trend: dict)

Subtract the previously calculated trend from (new) data. Only linear is implemented.

s2spy.preprocess._get_climatology(data: xarray.Dataset | xarray.DataArray, timescale: Literal[monthly, weekly, daily])

Calculate the climatology of timeseries data.

s2spy.preprocess._subtract_climatology(data: xarray.Dataset | xarray.DataArray, timescale: Literal[monthly, weekly, daily], climatology: xarray.Dataset | xarray.DataArray)

s2spy.preprocess._check_input_data(data: xarray.DataArray | xarray.Dataset)

Check the input data for compatiblity with the preprocessor.

Parameters:

data – Data to validate.

Raises:

• ValueError – If the input data is of the wrong type.

• ValueError – If the input data does not have a ‘time’ dimension.

s2spy.preprocess._check_temporal_resolution(timescale: Literal[monthly, weekly, daily]) → Literal[monthly, weekly, daily]

s2spy.preprocess._check_data_resolution_match(data: xarray.DataArray | xarray.Dataset, timescale: Literal[monthly, weekly, daily])

Check if the temporal resolution of input is the same as given timescale.

class s2spy.preprocess.Preprocessor(rolling_window_size: int | None, timescale: Literal[monthly, weekly, daily], rolling_min_periods: int = 1, subtract_climatology: bool = True, detrend: str | None = 'linear')

Preprocessor for s2s data.

property trend: dict

Return the stored trend (dictionary).

property climatology: xarray.DataArray | xarray.Dataset

Return the stored climatology data.

fit(data: xarray.DataArray | xarray.Dataset) → None

Fit this Preprocessor to input data.

Parameters:

data – Input data for fitting.

transform(data: xarray.DataArray | xarray.Dataset) → xarray.DataArray | xarray.Dataset

Apply the preprocessing steps to the input data.

Parameters:

data – Input data to perform preprocessing.

Returns:

Preprocessed data.

fit_transform(data: xarray.DataArray | xarray.Dataset) → xarray.DataArray | xarray.Dataset

Fit this Preprocessor to input data, and then apply the steps to the data.

Parameters:

data – Input data for fit and transform.

Returns:

Preprocessed data.