[Python] 강화학습 예제, CartPole-v1 step(다음 행동) 알아보기

https://github.com/ParkGyeongTae/dqn-atari-breakout/tree/main/1_cartpole-v1-example

GitHub - ParkGyeongTae/dqn-atari-breakout

 

### Step 1) 아나콘다 가상환경 만들기

- conda create -n py37 python=3.7

### Step 2) 가상환경 접속

- conda activate py37

### Step 3) 라이브러리 설치

- pip install gym==0.23.1
- pip install pygame==2.1.2

 

3_cartpole-v1-step.py

import gym
# import random

env = gym.make('CartPole-v1')

observation = env.reset()
action = env.action_space.sample()
# action = random.randrange(0, 2)
step = env.step(action)

print('observation : ', observation)
print('action      : ', action)
print('step        : ', step)

 

python 3_cartpole-v1-step.py

 

python 3_cartpole-v1-step.py
python 3_cartpole-v1-step.py
python 3_cartpole-v1-step.py

 

'''

import gym

env = gym.make('CartPole-v1')

# 에피소드 시작
observation = env.reset()

# 임의의 행동 실행
action = env.action_space.sample()

# step : action을 실행했을 때의 결과 (observation, reward, done, info)
step = env.step(action)

print('first_observation : ', first_observation)
print('action            : ', action)
print('step              : ', step)

# 결과
# first_observation :  [0.03486872  0.04418433 -0.00933439  0.02904002]
# action            :  0
# step              :  (array([ 0.0357524 , -0.15080252, -0.00875359,  0.3187633 ], dtype=float32), 1.0, False, {})

'''