search

FabianFuchsML / se3-transformer-public

code for the SE3 Transformers paper: https://arxiv.org/abs/2006.10503
475 stars 69 forks source link

Hyper-parameters of Nbody experiement #25

Open cermeng opened 2 years ago

cermeng commented 2 years ago

Hi Fabian,

I run the experiment of Nbody. However, with the default hyper-parameters and those recommended in README file, I got a different Test pos_mse from your paper. My result is around 0.03, and your result shown in paper is 0.0076. The difference of Test vel_mse also exists.

I am wondering if you can provide your hyper-parameter setting of Nbody experiment so that I can try to reproduce your result.

Thanks, Yunmeng

FabianFuchsML commented 2 years ago

Hi!

Great to hear that you are interested in our work! :)

My first thought is that you may have created the simulation with 20 particles (which I think is the default setting), but in the paper we used 5. Obviously, a system with 20 particles is more complex and will lead to larger errors.

Secondly, keep in mind that this is not the exact code from the paper but a reimplementation (IP reasons), both for the simulation and for the model. Hence, the results won't be exactly the same. However, if I do remember correctly, when we ran this code here with 5 particles, we got slightly better results than in the paper.

cermeng commented 2 years ago

Hi,

The particle number is indeed 20, I will run code again with 5 particles.

Thanks for your quick reply!