Problem with tutorial-1c

[lruba939]

Hi,

First of all, thank you so much — this software looks exactly like what I’ve been looking for. Great work!

I’m currently following the tutorial, but in section 1c I encountered a problem. Since it appears so early in the tutorial, I’m concerned it might affect later usage as well.

In a local .ipynb file (using VS Code), I get the following warnings:

/home/ruba/miniconda/envs/optilandTorch/lib/python3.11/site-packages/numpy/_core/function_base.py:162: RuntimeWarning: invalid value encountered in multiply
  y *= step
/home/ruba/miniconda/envs/optilandTorch/lib/python3.11/site-packages/numpy/_core/function_base.py:172: RuntimeWarning: invalid value encountered in add
  y += start
/home/ruba/miniconda/envs/optilandTorch/lib/python3.11/site-packages/optiland/geometries/standard.py:94: RuntimeWarning: invalid value encountered in sqrt
  self.radius * (1 + be.sqrt(1 - (1 + self.k) * r2 / self.radius**2))

I have tried two different Conda environments:

• one with Optiland installed from source (cloned from GitHub),
• one installed via:

pip install optiland[torch]

However, the result is the same in both cases.

Am I doing something wrong?

Additional information:

• Python==3.11
• numpy==2.3.5 (both envs)
• Ubuntu 22.04.5 LTS

Thanks in advance!

[HarrisonKramer]

Hi @lruba939,

Thanks for opening this! Good find. This is a bug - you are not doing anything wrong. The issue is minor and relates to aperture extent plotting. I fixed this just now in #487. I will merge that shortly.

Regards,
Kramer

[omni-front]

sounds like you're running into some issues with the calculations involving numpy and possibly the parameters being used. the warnings suggest that somewhere in your code, there's a chance that operations are trying to handle NaNs or infinite values due to invalid calculations like dividing by zero. double-check the inputs to your functions, especially any that might be passing unexpected values like infinity or NaN. harrisonkramer's response suggests there's already a fix in the works, so keep an eye on any updates or patches that might resolve this.