SpikeInterface params ar enot properly propagated

[alejoe91]

Hi @jamesjun

If I run:

sorter_params = {'fParfor': True, 'fGpu': False, 'freq_max': 3000}
sorting = ss.run_ironclust(rec, **sorter_params)

It seems that the fParfor and fGpu are not propagated to the matplab script. Here is a snippet of the output:

===================================================
IronClust Version: 2
===================================================
Removed 0 lock(s).
Running irc2.m (5.9.8)
Created /Users/abuccino/Documents/Codes/spike_sorting/spikeinterface/spikewidgets/tmp_ironclust/tmp/raw_geom.prm
Read /Users/abuccino/Documents/Codes/spike_sorting/spikeinterface/spikewidgets/tmp_ironclust/ironclust_dataset/raw.mda (1/4), took 0.9s (547.5 MB/s, 500.2 MB)
    get_prinvec_: took 0.1s
    Detecting 1/4: 27253 spikes found (4027.1 spikes/s, 73.9 MB/s, took 6.8 s)
    Detecting 2/4: 27334 spikes found (4690.8 spikes/s, 85.9 MB/s, took 5.8 s)
    Detecting 3/4: 27356 spikes found (4649.2 spikes/s, 85.0 MB/s, took 5.9 s)
    Detecting 4/4: 25472 spikes found (4699.4 spikes/s, 86.0 MB/s, took 5.4 s)
Detection took 24.2s and used NaN GiB (fParfor=0, fGpu=1)
Saving a struct to /Users/abuccino/Documents/Codes/spike_sorting/spikeinterface/spikewidgets/tmp_ironclust/tmp/detect_7e30aa58e1e86da80b67533268fbfdf0/detect_irc.mat: took 0.1s.
Clustering
Calculating drift similarity...
    took 0.0s
sort_page_: calculating Rho...
Page 1/1 C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.CC.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C. took 4.3s
calculating Rho took 4.3s
sort_page_: calculating Delta...
Page 1/1 C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C.C. took 2.4s
calculating Delta took 2.4s
sort_long_: took 6.7s (fGpu=1, fParfor=0)

Can you double check this? I tried to track it down but I couldn't. Thanks, Alessio

[alejoe91]

Note that other params are correctly set. I checked thatfreq_max=3000 is correctly set in matlab

[ferchaure]

This issue is related to #54