search

bn2302 / rl_torcs

Reinforcement learning with docker and torcs
20 stars 11 forks source link

bug fix! #1

Open zdx3578 opened 7 years ago

zdx3578 commented 7 years ago 
                if (done or episode_step_count >= max_episode_length):
                    break

step one then restarting

now can always running!

zdx3578 commented 7 years ago

apt-get install build-essential -y curl -O http://us.download.nvidia.com/XFree86/Linux-x86_64/375.51/NVIDIA-Linux-x86_64-375.51.run chmod +x ./NVIDIA-Linux-x86_64-.run ./NVIDIA-Linux-x86_64-.run -q -a -n -X -s rm ./NVIDIA-Linux-x86_64-*.run

this will change nvida cuda driver ; careful!! careful！！

bn2302 commented 7 years ago

Hi,

thanks for the info. Can you please send the suggested changes into a pull request?

Can you also please clarify, where you see the issue in the nvidia install script? It is meant for an AWS instance, so I assume that it starts from a clean ubuntu 16.04.

Best, Bastian

zdx3578 commented 7 years ago

@bn2302 when we complete ; we will send a pull request; thanks your work!

zdx3578 commented 7 years ago

https://github.com/ugo-nama-kun/gym_torcs/pull/3