search

Denys88 / rl_games

RL implementations
MIT License
800 stars 135 forks source link

Adding categorical cross_entropy loss to yaml config. #276

Open Eirikalb opened 4 months ago

Eirikalb commented 4 months ago

After reading the recent paper from Google Deepmind I am really tempted to try their methods out myself to benchmark some Online RL problems in Omniverse Isaac Gym. https://arxiv.org/abs/2403.03950.

I see that there is some mention of experimental support for it in the code but I also see in the changenotes that crossentropy loss has not been added to the yaml config yet. What is the easiest/ best path for me to try it out using this framework. All my current experiments are based on standard MSE loss, using standard network architectures.

Denys88 commented 4 months ago

Hi, I experimented with it here https://github.com/Denys88/rl_games/blob/cba782ceb772795628e52a3da3d5dc8c20ecb779/rl_games/algos_torch/network_builder.py#L171 yes it is not in config. I tested TwoHot encoding. Feel free to try more options. I can add it to the yaml config.