run: Compare other packages' functions for nonlinear least squares...

In ArkaB-DS/nlsCompare: Compare the best parameter values and sum-of-squares involved in solving nonlinear least squares (nls)

Description

Perform comparison of best reference values of parameters and sum of squares in a nonlinear problem with other packages' functions for nonlinear least squares.

Usage

run(machid, database, errorlog)

Arguments

machid	Takes the machine ID generated by machineId().
database	Takes the dataframe generated by create_db().
errorlog	Takes the dataframe generated by create_elog().

Details

The function first stores problems.csv and methods.csv into dataframes NLSproblems and NLSmethods respectively. It then sources the files, with names from NLSproblems, from the inst/scripts directory. That in turns creates NLSdata, NLSformula, NLSlower, NLSmethods, NLSproblems, NLSsubset, NLSupper, NLSweights, NLSstart, NLSssquares, NLSpars, refsol, NLStag, and NLSref. These along, with algorithm and control from NLSmethods are used inside create_runline, a helper function. The returned string is then used by the solver from NLSproblems to create nls models. From there, the parameter values and sum of squares are compared with the reference values that are present in "scripts" via problems.csv. Finally, the results are used to modify the argument dataframes database and errorlog and they are returned as a list.

Value

It returns a list of 2 items:

database	updated database dataframe
errorlog	updated error log dataframe

See Also

all.equal

ArkaB-DS/nlsCompare documentation built on Jan. 1, 2022, 2 p.m.