CartPole-v0.py

import numpy as np
import gym
from gym import wrappers
env = gym.make('CartPole-v0')

nsteps = 200

def run_episode(env, parameters):
    observation = env.reset()
    totalreward = 0
    for _ in range(nsteps):
        env.render()
        action = 0 if np.matmul(parameters, observation) < 0 else 1
        observation, reward, done, info = env.step(action)
        totalreward += reward
        if done:
            break
    env.close()
    return totalreward
  
bestparams = None
bestreward = 0

for _ in range(10000):
    parameters = np.random.rand(4) * 2 - 1
    tmpreward = run_episode(env, parameters)
    print(tmpreward)
    if tmpreward > bestreward:
        bestreward = tmpreward
        bestparams = parameters
        
        if tmpreward == nsteps:
            break