Errors outputting plots

[rpx91]

I get the nice pretty graphs output fine but the following errors are reported on the console

(venv) C:\Users\dushy\Impulcifer>python impulcifer.py --test_signal="data/sweep-6.15s-48000Hz-32bit-2.93Hz-24000Hz.pkl" --dir_path="data/my_hrir" --plot
 ** On entry to DLASCLS parameter number  4 had an illegal value
 ** On entry to DLASCLS parameter number  4 had an illegal value
Traceback (most recent call last):
  File "impulcifer.py", line 257, in <module>
    main(**create_cli())
  File "impulcifer.py", line 76, in main
    compensate_headphones(headphones, hrir, fig_path=fig_path)
  File "impulcifer.py", line 141, in compensate_headphones
    min_mean_error=False
  File "C:\Users\dushy\Impulcifer\venv\lib\site-packages\autoeq\frequency_response.py", line 1036, in compensate
    iterations=DEFAULT_TREBLE_SMOOTHING_ITERATIONS
  File "C:\Users\dushy\Impulcifer\venv\lib\site-packages\autoeq\frequency_response.py", line 1180, in smoothen_fractional_octave
    treble_f_upper=treble_f_upper
  File "C:\Users\dushy\Impulcifer\venv\lib\site-packages\autoeq\frequency_response.py", line 1137, in _smoothen_fractional_octave
    y_normal = savgol_filter(y_normal, self._window_size(window_size), 2)
  File "C:\Users\dushy\Impulcifer\venv\lib\site-packages\scipy\signal\_savitzky_golay.py", line 337, in savgol_filter
    coeffs = savgol_coeffs(window_length, polyorder, deriv=deriv, delta=delta)
  File "C:\Users\dushy\Impulcifer\venv\lib\site-packages\scipy\signal\_savitzky_golay.py", line 139, in savgol_coeffs
    coeffs, _, _, _ = lstsq(A, y)
  File "C:\Users\dushy\Impulcifer\venv\lib\site-packages\scipy\linalg\basic.py", line 1218, in lstsq
    raise LinAlgError("SVD did not converge in Linear Least Squares")
numpy.linalg.LinAlgError: SVD did not converge in Linear Least Squares

[jaakkopasanen]

Thanks for the issue. I'll look into it.

Could be share the whole folder with all the recordings etc? Maybe there is something that will trigger some edge case I haven't taken into account.

[rpx91]

Ah nuts - I overwrote it with new recordings. Will post one when it happens again.

As an aside - if recordings could be moved into a time stamped folder that'd be ideal for trouble shooting and tracing things back.

[jaakkopasanen]

Did the same happen again with the newer recordings or was it just the old ones?

I've taken a position of not trying to implement Windows explorer functionality in Impulcifer. Recordings can be moved to timestamped folders by creating the folders in Windows and moving the files there.

[rpx91]

Understood - in my experience of running multiple HRIR comps it's a good way to trouble shoot or figure out exactly why a recording resulted in a particularly good HRIR.

I'll figure out a script to do that.