Skip to content

A simple Reinforcement learning project that runs the well known Four Rooms as a test environment

Notifications You must be signed in to change notification settings

NivDeveloper/FourRooms

Repository files navigation

FourRooms

A simple Reinforcement learning project that runs the well known Four Rooms as a test environment

Scenarios

Scenario1 - reinforcement learning agent that finds one package in the environment

Scenario2 - reinforcement learning agent that fetches 3 packages from a grid environment.

Scenario3 - agent that finds 3 packages from an environment in the order of red, green, blue

Running

Running each of these programs requires only running the following command in the terminal

python3 ScenarioX.py

where X represents the number of the Scenario that you want to run.

the program will then begin to train the agent in the environment wich the specified amount of EPOCHS. the following message will display in the terminal while the agent is training

Training... numnber of epochs ( X / N )

where N represents the total number of epochs that will be trained and X shows the current epoch that the agent is training within.

About

A simple Reinforcement learning project that runs the well known Four Rooms as a test environment

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published