mgwr.gwr.GWR¶
-
GWR.
fit
(self, ini_params=None, tol=1e-05, max_iter=20, solve='iwls', lite=False, pool=None)[source]¶ Method that fits a model with a particular estimation routine.
- Parameters
- ini_betasarray, optional
k*1, initial coefficient values, including constant. Default is None, which calculates initial values during estimation.
- tol: float, optional
Tolerence for estimation convergence. Default is 1.0e-5.
- max_iterinteger, optional
Maximum number of iterations if convergence not achieved. Default is 20.
- solvestring, optional
Technique to solve MLE equations. Default is ‘iwls’, meaning iteratively ( re)weighted least squares.
- litebool, optional
Whether to estimate a lightweight GWR that computes the minimum diagnostics needed for bandwidth selection (could speed up bandwidth selection for GWR) or to estimate a full GWR. Default is False.
- poolA multiprocessing Pool object to enable parallel fitting; default is None.
- Returns
- :
If lite=False, return a GWRResult instance; otherwise, return a GWRResultLite instance.