Unable to allocate memory to load spike_times/spike cluster file from kilosort

[Andrianarivelo]

Hi,

I tried to visualize NPX 2.0 data after KS4 spike sorting.
I loaded them in rastermap gui using load spike_times and spike_cluster but I got a message saying
numpy.core._exceptions._ArrayMemoryError: Unable to allocate 284. GiB for an array with shape (282, 269997161) and data type float32

Is it a memory issue of my computer (I don't have 284GiB of RAM) ?
How can I fix this ?

Thanks

[jnjnnjzch]

This problem is because the spike_times.npy and spike_clusters.npy are assumed to be in seconds. However, the two files generated by kilosort count in samples.

And the easiest solution might be changing st_bin to st_bin * fs where fs is your sampling rate. For example, if you want time bin to be 100 millisecond, and your sample is fs=30000, then you should set st_bin to 100*30000=3000000.