sac_learn --train was crash

deep-reinforcement-learning-book / Chapter16-Robot-Learning-in-Simulation

Chapter 16 Robot Learning in Simulation in book Deep Reinforcement Learning: example of Sawyer robot learning to reach the target with paralleled Soft Actor-Critic (SAC) algorithm, using PyRep for Sawyer robot simulation and game building. The environment is wrapped into OpenAI Gym format.

https://deep-reinforcement-learning-book.github.io

49 stars 19 forks source link

sac_learn --train was crash #1

Closed leeivan1007 closed 4 years ago

leeivan1007 commented 4 years ago

Package version： V-REP 3.6.2 torch 1.4.0 It will be crash when I run the train mode.

leeivan1007 commented 4 years ago

I had changed the version of torch. 1.4->1.2 but it is the same outcome. Version: torch:1.2.0 gym:0.15.4 v-rep:3.6.2

this is the other infomation that I stop the progress: 2020-04-15 12-16-12 的螢幕擷圖

quantumiracle commented 4 years ago

Hi, Could you please provide more information or log details when the program crashed? It seems several processes have been called, so is it caused by overuse of memory or GPUs? Zihan

leeivan1007 commented 4 years ago

I find the problem that the GPU memory is no enough. when I close the web browser, the problem is saved. Think you!