gfdlvitals.extensions module

Pandas class extension for gfdlvitals

class gfdlvitals.extensions.Timeseries(f, var, scale=1.0, multiply_by_area=False, legacy_land=False, start=None, end=None)

Bases: object

Timeseries class object

Parameters:

f (str, path-like) – Input SQLite file
var (str) – Variable to extract
scale (float, optional) – Scale data by this factor, by default 1.0
multiply_by_area (bool, optional) – Multiply variable by cell area before returning, by default False
legacy_land (bool, optional) – Read legacy version of the land SQLite files, by default False
start (int, optional) – Specify start year, by default None
end (int, optional) – Specify end year, by default None

property data

property t

class gfdlvitals.extensions.VitalsDataFrame(data=None, index: Axes | None = None, columns: Axes | None = None, dtype: Dtype | None = None, copy: bool | None = None)

Bases: DataFrame

Pandas class extension for holding the Vitals data

Parameters:: pd (pandas.DataFrame) – Input Pandas DataFrame object
Returns:: Class extension object
Return type:: gfdlvitals.VitalsDataFrame

areasum(): Returns the area integrated variable based on the cell_measure method

build_netrad_toa()

Constructs netrad_toa from component terms if available

Parameters:: self (VitalsDataFrame)
Return type:: VitalsDataFrame

detrend(reference=None, order=1, anomaly=True, return_coefs=False)

Detrend VitalsDataFrame object

Parameters:

reference (gfdlvitals.DataFrame, optional) – Reference VitalsDataFrame, by default None
order (int, optional) – Polynomial order to use for fitting, by default 1
anomaly (bool, optional) – Results as anomalies from the fit, by default True
return_coefs (bool, optional) – Return polynomial fit coeffiecients, by default False

Returns:

Extended dataset

Return type:

self

extend(maxlen)

Extend VitalsDataFrame to a set length and pad with NaNs

Parameters:: maxlen (int) – New length of the VitalsDataFrame
Returns:: Extended dataset
Return type:: self

smooth(window, extrap=False)

Apply a smoother to the dataset

Parameters:

window (int) – Smoothing filter length
extrap (bool, optional) – Extrapolate data on the ends, by default False

Returns:

Smoothed dataset

Return type:

self

trend(order=1)

Fits a trend to the VitalsDataFrame object

Parameters:: order (int, optional) – Polynomial order to use for fitting, by default 1
Returns:: Fitted trend dataset
Return type:: self

ttest(df2)

Performs t-test between two instances of VitalsDataFrame

Parameters:: df2 (VitalsDataFrame) – Comparison data set
Returns:: Contains p-values for variables common between the two VitalsDataFrames
Return type:: pandas.DataFrame

gfdlvitals.extensions.open_db(dbfile, variables=None, yearshift=0.0, legacy_land=False, start=None, end=None): Function to read sqlite dbfile

gfdlvitals.extensions.reformat_time_axis(ax=None)

Reformats x-axis labels to YYYY format

Parameters:: ax (matplotlob.pyplot.figure.axis, optional) – axis object to be reformatted, by default None

gfdlvitals.extensions.ttest_ind_auto(arr1, arr2, axis=0)

Performs a t-test that adjusts the degrees of freedom based on the autocorrelation of the dataset. See Krasting et al. 2013 for more details DOI: 10.1175/JCLI-D-12-00832.1

Parameters:

arr1 (numpy.ndarray) – First array of data points
arr2 (numpy.ndarray) – Second array of data points
axis (int, optional) – Axis to perform t-test, by default 0

Returns:

t-statistic, probability, arr1 autocorrelation, arr2 autocorrelation

Return type:

tuple