Fast lyapunov index calculation

[Daisy0302hhh]

Hi there, I am very interested in the research content of this article,and I want to express my appreciation for making code open source.

I notice that this article uses FLI as a chaos indicator. When solving the FLI of the corresponding solution for an initial value, I first combine the original equation and the variational equation to solve with odeintw, where v(0) is randomly selected some vectors with module length 1 but different directions. Secondly, the exponent_v0 is denoted as lyapunov_exponent_v0 by taking the upper bound of the modular length of v(t) obtained from each initial value v(0) in the 0-T time range. Finally, with respect to different initial values v(0), the final lyapunov_exponent is obtained.

According to the above method and the threshold value in the paper, the test results on H_2, eps=0.02 are obtained as shown in the figure below.

Do you have any suggestions on how I might address this issue? Any guidance or assistance you can provide would be greatly appreciated.

Thanks again for your invaluable support!

Cheers!

[vrodriguezf]

Hi,

Thank you for your interest in our work! What's your question exactly? I can help you with the ML side of the paper and the usage of the code. For domain specific questions please reach out prof. Celleti, Gales or Vasile.

Best! Victor

[Daisy0302hhh]

Hi!

Thanks for your response!

Sorry, I didn't describe my problem clearly.Let me simplify my question as follows.

I noticed that FLI was used as a chaos indicator for testing in the article. Specifically, what is the algorithm for calculating FLI? Because I wrote the code in accordance with the description in the article and the threshold Settings, I did not get the same result as in fig3(b) (my result is the image displayed on github). I don't know what the reason is.

Thanks again for your invaluable support!

Best Hjy

------------------ 原始邮件 ------------------ 发件人: "vrodriguezf/mlchaos" @.>; 发送时间: 2024年9月28日(星期六) 下午4:55 @.>; @.**@.>; 主题: Re: [vrodriguezf/mlchaos] Fast lyapunov index calculation (Issue #2)

Hi,

Thank you for your interest in our work! What's your question exactly? I can help you with the ML side of the paper and the usage of the code. For domain specific questions please reach out prof. Celleti, Gales or Vasile.

Best! Victor

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>