python - Evaluating a Gaussian Fit -
i'd know ways determine how gaussian function fitting data.
here few plots i've been testing methods against. currently, i'm using rmse of fit versus sample (red fit, blue sample).
for instance, here 2 fits:


and here 2 terrible fits should flagged bad data:

in general, i'm looking suggestions of additional metrics measure goodness of fit. additionally, can see in second 'good' fit, there can other peaks outside data. currently, these penalized rsme method, though should not be.
i'm looking suggestions of additional metrics measure goodness of fit.
the one-sample kolmogorov-smirnov (ks) test starting point.
i'd suggest wikipedia article introduction.
the test available in scipy scipy.stats.kstest. function computes , returns both ks test statistic , p-value.
Comments
Post a Comment