# Our goal is to keep the pole upright as long as possible until score of 500
# the more time the more score
for time in range(700):
env.render()
# Decide action
action = agent.act(state) # maximum action ; pass our vector state to our NN in which we have state_size neurons
# Advance the game to the next frame based on the action.
# Reward is 1 for every frame the pole survived
next_state, reward, done, _ = env.step(action)
reward = reward if not done else -10
# we are turning our next_state into a one dimensional matrix which is a vector
# to calculate the maximum future reward for next state ; cause our model input
# is a one dimensional matrix which is a vector in which in our case is 4 neurons
next_state = np.reshape(next_state, [1, state_size])
# Remember the previous state, action, reward, and done
agent.remember(state, action, reward, next_state, done)
# make next_state the new current state for the next frame.
state = next_state
# done becomes True when the game ends
# ex) The agent drops the pole
if done:
agent.update_target_model()
print("episode: {}/{}, score: {}, e: {:.2}"
.format(e, EPISODES, time, agent.epsilon))
break
if len(agent.memory) > batch_size:
# train the agent with the experience of the episode
# loss = agent.replay(batch_size)
agent.replay(batch_size)
# Logging training loss every 10 timesteps
# if time % 10 == 0:
