minerEnv.step(
str(action)
) # Performing the action in order to obtain the new state
reward = minerEnv.get_reward() # Getting a reward
new_state_map, new_state_users = minerEnv.get_state(
) # Getting a new state
terminate = minerEnv.check_terminate(
) # Checking the end status of the episode
#t1=0
#t2=0
#t3=0
# Add this transition to the memory batch
#tmp_t1 = time.time()
memory.append(state_map, state_users, action, reward,
new_state_map, new_state_users, terminate)
#t1 = time.time() - tmp_t1
episode_reward += reward # Plus the reward to the total reward of the episode
state_map = new_state_map # Assign the next state for the next step.
state_users = new_state_users # Assign the next state for the next step.
score = minerEnv.state.score
"""
summary_2 = tf.Summary()
summary_2.value.add(tag='time_append', simple_value=t1)
summary_2.value.add(tag='time_take_samples', simple_value=t2)
summary_2.value.add(tag='time_train', simple_value=t3)
summary_writer_time.add_summary(summary_2, total_step)
summary_writer_time.flush()
"""
