brendonw1
commented
6 years ago I agree with this and thought I did something to address this already. But regardless glad you're dong it.
On Thu, May 17, 2018 at 6:27 PM, Dan Levenstein notifications@github.com wrote:
Some things fail if you make your sessionInfo file before spike sorting. Your spike groups become: spikeGroups.nSamples = []; spikeGroups.groups = {}; spikeGroups.nGroups = number of anatomical groups
Which means that when you load bz_getSessionInfo to put in regions, it fails because it tries to load the first element of spikeGroups.groups etc etc.
We clearly don't want to define spike groups before spike sorting, but it would be good to still be able to use anatomical groups... for the purposes of defining channel regions etc. Should the default in bz_getSessionInfo etc to check for spikeGroups and if there are none then use the ElecGp field in the sessionInfo?
Either way, we should change the nGroups field to 0, because number of anatomical groups is not always 0.
This is the issue of group redundancy that is "killing @DavidTingley https://github.com/DavidTingley slowly" - (see line 139 of LoadParameters).
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/buzsakilab/buzcode/issues/197, or mute the thread https://github.com/notifications/unsubscribe-auth/ADXrTbWJsW_YGTfrIHXEeRHTru08McdIks5tzflcgaJpZM4UD3Wo .
dlevenstein
Some things fail if you make your sessionInfo file before spike sorting. Your spike groups become: spikeGroups.nSamples = []; spikeGroups.groups = {}; spikeGroups.nGroups = number of anatomical groups
Which means that when you load bz_getSessionInfo to put in regions, it fails because it tries to load the first element of spikeGroups.groups etc etc.
We clearly don't want to define spike groups before spike sorting, but it would be good to still be able to use anatomical groups... for the purposes of defining channel regions etc. Should the default in bz_getSessionInfo etc to check for spikeGroups and if there are none then use the ElecGp field in the sessionInfo?
Either way, we should change the nGroups field to 0, because number of anatomical groups is not always 0.
This is the issue of group redundancy that is "killing @DavidTingley slowly" - (see line 139 of LoadParameters).