miketrumpis

Pretty sure I've only seen it on Ubuntu 20.04.4. Correct me if I'm wrong, I believe the 0.94 version only uses spawning for multiprocessing. The stalling scenario is either using...

I will try to run under the git main soon and play with the multiprocessing context. I have a list of jobs that have failed, but not 100% sure the...

> @alejoe91 : we do not use loky. It was too bugy. The ProcessPoolExecutor is in python core. I was taking another look at this too--so I presume that uses...

I thought I'd weigh in here (hi @arokem !).. the problem is with computing t-distributed confidence intervals, scaled by the jackknife standard error: ```python t025_limit = coh_mat_xform + dist.t.ppf(.025, K...

On second read.. now that I'm looking closer, the script is already doing the normalize / un-normalize steps. @Xunius are you saying the upper/lower CIs are *still* outside the [0,...

edit: disregard, bad idea.. I'll give this some more time Good point! Maybe offset the basis? ```python def coh_to_norm(x): return np.arctanh((x + 1) / 2) def norm_to_coh(y): return (np.tanh(y) +...

Those transforms weren't quite right. This appears to handle the tails appropriately (see [gist](https://gist.github.com/miketrumpis/f12857b82642209fae46a9b4fd24f439)) Also, these are simplified (not scaled by DoF) ```python def coh_to_norm(x): return np.arctanh(2 * x -...

I'm seeing eq 24 as complex, but I agree: it does look like the 2007 sig proc magazine paper is suggesting to use the arctanh transform of the modulus of...

Apologies for the lack of transparency here, as this is not public data. I've shared a description of the problem in the fork I made. My intention was to PR,...

After thinking some more, I'm not sure it's safe to assume that the main electrode table has metadata that applies to all derived `ElectricalSeries`. As I understand, there is only...