reacher_pybulletgym.py

import gym
import pybulletgym

env = gym.make('ReacherPyBulletEnv-v0')
env.render(mode="human")

for episode in range(20):
    state = env.reset()             
    rewards = []
    
    for t in range(30000):
        env.render()                 
        # action = env.action_space.sample()  
        action = [0.00001, 0]
        print(action)
        state, reward, done, info = env.step(action) 
        rewards.append(reward)
        
    cumulative_reward = sum(rewards)
    print("episode {} | cumulative reward : {}".format(episode, cumulative_reward))  
    
env.close()