vis is an open source Python library for developing and comparing reinforcement learning algorithms by providing a standard API to communicate between learning algorithms and environments, as well as a standard set of environments compliant with that API

We are support Python 3.7, 3.8, 3.9, 3.10 on Linux and Windows
To install the base vis library, use:

• Using pip:

  Update later

• Directly from source (recommended):

  git clone https://github.com/ngoxuanphong/vis.git
  cd vis
  pip install -r requirements.txt

import env

env.make()
# Run game
print(env.run())

# Visualization one game
env.render(Agent = env.Agent_random)

#Play with bot in game
env.render(Agent = 'human')

import env
from numba import njit
import numpy as np

@njit()
def Agent(state, agent_data):
    validActions = env.getValidActions(state)
    actions = np.where(validActions==1)[0]
    action = np.random.choice(actions)
    return action, agent_data
    
env.make()
env.run(Agent, num_game = 1000, agent_model = [0], level = 0)

env.render(Agent = Agent)

Example

Please refer to Wiki for complete usage details

vis includes 20 games, which are listed below along with the number of actions, observations, time run, win rate they have:

Please refer to Wiki for more details.