[BUG] error in get_half_of_the_data()

[HDClark94]

The bug:

within PostSorting,
within compare_first_and_second_half.py,
within get_half_of_the_data_cell(prm, spike_data_in, synced_spatial_data_in, half='first_half'):

spike_data_half = None
end_of_first_half_seconds = (synced_spatial_data.synced_time.max() - synced_spatial_data.synced_time.min()) / 2
end_of_first_half_ephys_sampling_points = end_of_first_half_seconds * prm.get_sampling_rate()

line 2 looks to calculate the midpoint of the recording in seconds however instead computes half the range. for example:

min=2 seconds
max=5 seconds
then end_of_first_half_seconds should be 3.5

using the above equation
end_of_first_half_seconds = 1.5
this is less that the minimum time and therefore not possible, a fix I've implemented on my dual sorting branch uses

spike_data_half = None
end_of_first_half_seconds = (synced_spatial_data.synced_time.max() + synced_spatial_data.synced_time.min()) / 2
end_of_first_half_ephys_sampling_points = end_of_first_half_seconds * prm.get_sampling_rate()

swapping the minus for a plus changes the computation to taking the midpoint of max and min.

Could someone please check to confirm my interpretation on this?

Running environment

• dual_sorting branch
• Eleanor instance

[klaragerlei]

@HDClark94 I think I only used this where min was 0 and didn't notice, sorry.

I think your interpretation is correct, (but maybe wait for one more person to check).

[4iar]

@HDClark94 I think this makes sense - nice find!

The function get_half_of_the_data() also uses the same calculation (here), would that need changing too?

[klaragerlei]

@HDClark94 this is fixed now, right?