MLE tomography is exceeding max func evals in scipy.optimize.leastsq

[dannypaulson]

Hi there,

I'm a new user of your package and wanted to ask a question about an issue I've encountered. For some relatively simple tomography examples on 1 or 2 qubits, I am getting warnings from scipy.optimize.leastsq that the maximum number of function evaluations has been exceeded, meaning the optimization exited before it converged. The leastsq function has a parameter maxfeval that can be used to set the number of allowed evaluations to some higher number to let the optimization converge, but this parameter is not exposed in the StateTomography method of the Tomography class or anywhere else that I can see. Do you see a fix for this?

Thanks for your help!

Danny

[smturro2]

Hi Danny, Thanks for reaching out. I think this is a wonderful idea. We currently do not have anyone maintaining the code to implement this feature. I’m going to see if I still have permissions to implement the request. Progress may be slow but I’ll keep you posted. Thanks, Scott Turro

…

On Tue, Jan 3, 2023 at 5:06 PM Danny Paulson ***@***.***> wrote: Hi there, I'm a new user of your package and wanted to ask a question about an issue I've encountered. For some relatively simple tomography examples on 1 or 2 qubits, I am getting warnings from scipy.optimize.leastsq that the maximum number of function evaluations has been exceeded, meaning the optimization exited before it converged. The leastsq function has a parameter maxfeval that can be used to set the number of allowed evaluations to some higher number to let the optimization converge, but this parameter is not exposed in the StateTomography method of the Tomography class or anywhere else that I can see. Do you see a fix for this? Thanks for your help! Danny — Reply to this email directly, view it on GitHub <#29>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKH5GXF2PWBNAJYKU5NH65LWQSWHVANCNFSM6AAAAAATQFSJCE> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[smturro2]

@dannypaulson A new version has been published. version 1.05.0 will allow you to pass the parameters in through the config

[dannypaulson]

Thanks so much!