Hi there! Thanks for publicizing such an awesome project!
I would like to ask how I can get the results of model evaluation similar to the results shown in the paper? Because it seems like only the training process is in the code right now.
Thanks!
The training code also performs evaluation so you can just run the training script till convergence and it will log the success rate to wandb along with videos of the agent performance.
Hi there! Thanks for publicizing such an awesome project!
I would like to ask how I can get the results of model evaluation similar to the results shown in the paper? Because it seems like only the training process is in the code right now.