problems with binspikes and estfr

[nirum]

• estfr needs to use filtfilt instead of lfilter
• both binspikes and estfr try and access a variable called cell that does not exist
• returned bins should be centered, not left bins

[nirum]

There are still problems with estfr-

the scipy.signal filter functions (lfilter and filtfilt) take in arguments given as coefficients in the numerator / denominator of a filter specification. Currently, only the gaussian profile is passed as an argument, which means filtering is not happening correctly?

I could be wrong, but I think we should go back to the way I was doing things before with creating a butterworth filter and using filtfilt or sticking with a gaussian but then using convolution.

documentation on filtfilt: http://docs.scipy.org/doc/scipy/reference/generated/scipy.signal.filtfilt.html

[nirum]

I went a bit further - and made sure that the array that is returned from the convolution is aligned to the original bspk array. That way peaks in bspk and the estimated firing rate are aligned.