You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I wanted to ask if the enforce_complex=False functionality is obsolete in the newer/newest versions.
With version 1.4.x if I set it to False, it will not do the imaginary part of the integration (with Vegas).
Whereas, in the 1.5.x version, it seems like whether you set it to True/False, it always does the imaginary part of the integration, although in the integrand there is no such imaginary part.
Is this functionality doing what it should do?
PS. The imaginary integration is fast of course as there is nothing to integrate and prints a 0 (x e-19 I) (still it is using computing resources).
Kind regards,
GD
The text was updated successfully, but these errors were encountered:
I think that the option that works in newer pySecDec versions is called contour_deformation. Setting it to False effectively
disables the calculation of the imaginary part.
I can confirm that the flag enforce_complex is not working as intendended. This should only have a minor impact on the integration time, but we will try to fix this.
Hi,
I wanted to ask if the enforce_complex=False functionality is obsolete in the newer/newest versions.
With version 1.4.x if I set it to False, it will not do the imaginary part of the integration (with Vegas).
Whereas, in the 1.5.x version, it seems like whether you set it to True/False, it always does the imaginary part of the integration, although in the integrand there is no such imaginary part.
Is this functionality doing what it should do?
PS. The imaginary integration is fast of course as there is nothing to integrate and prints a 0 (x e-19 I) (still it is using computing resources).
Kind regards,
GD
The text was updated successfully, but these errors were encountered: