n_actions = len(env.get_action_meanings())
print(n_actions)
action_noise = NormalActionNoise(mean=np.zeros(n_actions),
sigma=0.1 * np.ones(n_actions))
model = DDPG('MlpPolicy', env, action_noise=action_noise, verbose=1)
# model = DDPG('MlpPolicy', env, verbose=1)
model.learn(total_timesteps=10000, log_interval=10)
#model.save("ddpg_pendulum")
#env = model.get_env()
#del model # remove to demonstrate saving and loading
#model = DDPG.load("ddpg_pendulum")
score = 0
obs = env.reset()
while True:
action, _states = model.predict(obs)
obs, rewards, dones, info = env.step(action)
env.render()
score = score + 1
print(dones)
if dones:
# obs = env.reset()
print('finished', score)
break
#
# model.load('DDPG_test_2_SOC_point5_two_states')
mean_reward, std_reward = evaluate_policy(model, model.get_env(), n_eval_episodes=10)
print("Mean Reward = ", mean_reward)
epsi_sp_list = []
action_list = []
soc_list = []
Concentration_list = []
Concentration_list1 = []
obs = env.reset()
for _ in range(3600):
action, _states = model.predict(obs, deterministic=True)
obs, rewards, done, info = env.step(action)
epsi_sp_list.append(env.epsi_sp.item(0))
# Concentration_list.append(env.state_output['yp'].item())
# Concentration_list.append(env.state_output['yn'].item())
soc_list.append(env.state_of_charge.item())
action_list.append(action)
if done:
break
# obs = env.reset()
plt.figure()
plt.plot(soc_list)