A fork of Kyle LePage's spike field association code.
The parent of this source is the code provided on Kyle LePage's website, dated 20 Nov 2012.
Run get_cases_param_est_with_exp_many_f.m
The .m files: glm_sfa_*_chk_phase.m produce 2-page postscript files with diagnostics.
The first page is a plot of the filter transfer function and the second plot (right most column) has a plot of instantaneous frequency departure.
These plots should be "green everywhere" except at the trial boundaries. If the n_trim parameter is mis-specified you will see a big departure from this behaviour; or if there is a problem with "leakage from adjacent frequencies".