bitsauce / Carla-ppo

This repository hosts a customized PPO based agent for Carla. The goal of this project is to make it easier to interact with and experiment in Carla with reinforcement learning based agents -- this, by wrapping Carla in a gym like environment that can handle custom reward functions, custom debug output, etc.
MIT License
229 stars 56 forks source link

Low Efficiency on Data Collection #28

Open Ezharjan opened 8 months ago

Ezharjan commented 8 months ago

While running the data collector Python script in this repository, the efficiency is too low that it dropped to 1FPS even on a GPU3090 machine. Is this the problem of the ways collecting the data or the problem of Carla itself?