How to plot the convergence?

[brett18618]

Thank you for your work first. Could you tell me is there a method to plot the convergence with x axis is iteration numbers and y is the fitness function values?

[firefly-cpp]

Hi @brett18618 ,

Thanks!

Convergence plot is captured in this example. However, it is tailored to the nFES.

EXAMPLE

Do you prefer number of iterations on x?

[brett18618]

Thank you for your reply！I want to observe the convergence curve as the iteration number changed, is it possible to do that? Regards Brett

…

On Fri, Aug 2, 2019 at 3:26 PM firefly-cpp ***@***.***> wrote: Hi @brett18618 <https://github.com/brett18618> , Thanks! Convergence plot is captured in this example. However, it is tailored to the nFES. EXAMPLE <https://github.com/NiaOrg/NiaPy/blob/master/examples/run_pso.py> Do you prefer number of iterations on x? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#220?email_source=notifications&email_token=AKQYYBMBBVJDQTJ35YQ7VETQCQ7Z7A5CNFSM4II4TNXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3N4XHI#issuecomment-517720989>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AKQYYBIKNB4NNUOFFUS5SQTQCQ7Z7ANCNFSM4II4TNXA> .

[firefly-cpp]

Yes. It is possible.

We have to add a new method. Please give us a couple of days to complete this issue.

[brett18618]

Many thanks! Do you mind telling me when it is completed ?

…

On Fri, Aug 2, 2019 at 3:42 PM firefly-cpp ***@***.***> wrote: Yes. It is possible. We have to add a new method. Please give us a couple of days to complete this issue. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#220?email_source=notifications&email_token=AKQYYBJTQF5XD7FTUW3G6C3QCRBVDA5CNFSM4II4TNXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3N6EAI#issuecomment-517726721>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AKQYYBLMLX6Z3A527SEBWUDQCRBVDANCNFSM4II4TNXA> .

[firefly-cpp]

@brett18618 - Yes.

[brett18618]

And by the way, I was thinking each iteration corresponds one time evaluation which is actually the fitness function. It seemed I was wrong because when I set 100 iterations, the evaluations returned were only 58. Would you like to explain me the relationship between the number of evaluation and that of iteration?

…

On Fri, Aug 2, 2019 at 3:51 PM firefly-cpp ***@***.***> wrote: @brett18618 <https://github.com/brett18618> - Yes. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#220?email_source=notifications&email_token=AKQYYBLSLS6BY7ASLXRHB43QCRCYVA5CNFSM4II4TNXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3N7AGY#issuecomment-517730331>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AKQYYBMG732KBPW4YOKFLK3QCRCYVANCNFSM4II4TNXA> .

[firefly-cpp]

nFES counts each call of fitness function.

[brett18618]

So does that mean the fitness function would not be called in each iteration?

…

On Fri, Aug 2, 2019 at 4:14 PM firefly-cpp ***@***.***> wrote: nFES counts each call of fitness function. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#220?email_source=notifications&email_token=AKQYYBJPQW6WKUHDMJDNGHTQCRFOPA5CNFSM4II4TNXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3OBCCA#issuecomment-517738760>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AKQYYBKPWLSUXB2DJG7U6MTQCRFOPANCNFSM4II4TNXA> .

[GregaVrbancic]

@brett18618 we have added functionality (function return_conv) to the StoppingTask, which enables the retrieval of values for x and y axis for plotting the covariance graph.

[GregaVrbancic]

@all-contributors add @brett18618 for question

[allcontributors]

@GregaVrbancic

I've put up a pull request to add @brett18618! 🎉