drl-supply-chain-hilo-gym.py

import gym
from gym.spaces import Discrete, Box

class HiLoPricingEnv(gym.Env):
    def __init__(self, config):
        self.reset()
        self.action_space = Discrete(len(price_grid))
        self.observation_space = Box(0, 10000, shape=(2*T, ), dtype=np.float32)

    def reset(self):
      self.state = env_intial_state()
      self.t = 0
      return self.state

    # Returns next state, reward, and end-of-the-episode flag  
    def step(self, action):
      next_state, reward = env_step(self.t, state, action)
      self.t += 1
      self.state = next_state
      return next_state, reward, self.t == T - 1, {}