这是indexloc提供的服务,不要输入任何密码
Skip to content

rcognita is a flexibly configurable framework for agent-enviroment simulation with a menu of predictive and safe reinforcement learning controllers

License

Notifications You must be signed in to change notification settings

MusubaPy/rcognita

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation



Rcognita is a framework for DP and RL algorithm development, testing, and simulation.

Installation:

  • clone repo
  • run in terminal python3 setup.py install

Example of terminal comand:

python3 main_3wrobot.py --mode 5 --ndots 20 --radius 5 --dt 0.05

Arguments for comand line for single run:

  • mode - default 3
  • dt - default 0.05
  • Nactor - default 6
  • pred_step_size - default 5
  • init_x - used only if one point required and only with init_y and init_alpha default None
  • init_y - used only if one point required and only with init_x and init_alpha default None
  • init_alpha - could be used separetely for one point (without init_x and init_y) default None
  • ndots - number of dots for simulation default 25
  • radius - default 5
  • folder - default None will be created folder with name of current hour
  • is_log_data - saving date in csv file in folder data/date/hour(or_name) default True
  • is_print_sim_step - printing info about each step in terminal default False
  • is_visualization - will visualize interface of rcognita and all process default False

About

rcognita is a flexibly configurable framework for agent-enviroment simulation with a menu of predictive and safe reinforcement learning controllers

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 99.3%
  • Other 0.7%