geocontour.tests.timing module

Functions for comparatively testing speed of geocontour implementations

geocontour.tests.timing.masksearch(numtests=10, runspertest=1, boundname='generic_boundary', spacing=[1.75, 1.5, 1.25, 1, 0.75, 0.5, 0.25, 0.2], stat='min', plot=True, logax=False)

Test the timing of all mask search functions using timeit

Parameters:

numtests (int, default=10) – number of tests to run
runspertest (int, default=1) – number of function calls per test
boundname (str, default='generic_boundary') – the name of the boundary file to use (see geocontour/geocontour/data/)
spacing (array_like, default=[1.75,1.5,1.25,1,0.75,0.5,0.25,0.2]) – grid spacing value(s) to use for testing [i.e. if a list or numpy array is provided, testing will be run and results will populate for every value]
stat ({'mean', 'median', 'min', 'max'}, default='min') – select the statistical method for combining results from all tests for a given function
plot (bool, default=True) – select to plot results
logax (bool, default=False) – select to use log axes (if plotting)

Returns:

results – 2D Nx6 array of timing results in units of seconds where N is equal to len(spacing)

column1: number of cells searched

column2: center() timing

column3: center2() timing

column4: nodes() timing

column5: nodes2() timing

column6: area() timing

Return type:

ndarray

See also

contourtrace, geocontour.masksearch

Notes

The numtests parameter determines the number of tests that are later combined via the stat parameter. The runspertest parameter determines the number of times a function will be called in a given test. For example, if numtests = 10 and runspertest = 5, 50 total function calls will be made and result in 10 test results of the total timing of 5 consecutive function calls.

geocontour.tests.timing.contourtrace(numtests=10, runspertest=1, boundname='generic_boundary', spacing=[1.75, 1.5, 1.25, 1, 0.75, 0.5, 0.25, 0.2, 0.15, 0.125, 0.1, 0.09, 0.08, 0.07, 0.065, 0.06], stat='min', plot=True, logax=False)

Test the timing of all contour trace functions using timeit

Parameters:

numtests (int, default=10) – number of tests to run
runspertest (int, default=1) – number of function calls per test
boundname (str, default='generic_boundary') – the name of the boundary file to use (see geocontour/geocontour/data/)
spacing (array_like, default=[1.75,1.5,1.25,1,0.75,0.5,0.25,0.2]) – grid spacing value(s) to use for testing [i.e. if a list or numpy array is provided, testing will be run and results will populate for every value]
stat ({'mean', 'median', 'min', 'max'}, default='min') – select the statistical method for combining results from all tests for a given function
plot (bool, default=True) – select to plot results
logax (bool, default=False) – select to use log axes (if plotting)

Returns:

results – 2D Nx9 array of timing results in units of seconds where N is equal to len(spacing)

column1: number of cells in mask

column2: square() timing

column3: moore() timing

column4: moore_imp() timing

column5: pavlidis() timing

column6: pavlidis_imp() timing

column7: MSBF() timing

column8: ISBF() timing

column9: TSR() timing

Return type:

ndarray