search

mihdalal / raps

[NeurIPS 2021] PyTorch Code for Accelerating Robotic Reinforcement Learning with Parameterized Action Primitives
https://mihdalal.github.io/raps/
MIT License
72 stars 12 forks source link

Hardware and time used for training? #12

Closed BrightMoonStar closed 7 months ago

BrightMoonStar commented 7 months ago

Hi~ Could you share which hardware setup you used for training and roughly how long one training run took on that setup? Thanks a lot !

mihdalal commented 7 months ago

Hi, Thank you for your interest in our paper! From Section 4 of the paper: "To ensure consistency, we evaluate all methods on a single RTX 2080 GPU with 10 CPUs and 50GB of memory." For the length of training runs - it depends on the environment (see the plots wrt to wall clock time) but the max we ran any experiment for was 3 days.

BrightMoonStar commented 7 months ago

Oh, sorry, I miss it. Thank you !