Riesz s-Energy returns different outputs for same random seed

[LucasAlegre]

If you start two different Python processes and run:

get_reference_directions("energy", 4, 32, seed=42)

I get different outputs each time. I believe there is something wrong with how the random seed is set.

[blankjul]

very interesting. I quickly looked into it and yes I can reproduce it that even though the random seed is provided the results don't match.

However, the np.random.seed(seed) method seems to be called in all runs.

Need to look a little more into this and will let you know.

[blankjul]

I think the commit above fixes this now. Can you please check?

[LucasAlegre]

I cannot install it from source because I'm using python 3.8 and pymoo now requires python>=3.9 :/

[blankjul]

I will close this issue for now. You can use conda do update your environment if needed. Please reopen the issue if you find the current version does not resolve it.

[LucasAlegre]

Do you have an estimate of when a new release will be on pypi with this fix? Thanks!

[blankjul]

I usually try to create a release 2-3 times a year. Depending on the number of commits.

Was the fix successful with the current dev version?

[LucasAlegre]

When following the instructions in the README to install it from source, it installs version 0.6.0.1 and then the fix does not seem to have worked.