def train(self, episodes=1000, max_steps=800, plot_rewards=True):
# Initialize target network weights
self.actor.update_target_model(copy=True)
self.critic.update_target_model(copy=True)
scores, steps = np.empty(episodes), np.empty(episodes)
start = time.time()
for e in range(episodes):
score, step = self.run_episode(max_steps)
scores[e], steps[e] = score, step
print("Episode:", e, " steps:", step, " score:", score,
" time:",
time.time() - start)
ensure_saved_models_dir()
if plot_rewards:
t_time = time.time() - start
print("Mean score:", np.mean(scores), " Total steps:",
np.sum(steps), " total time:", t_time)
plot(scores)
plot_running_avg(scores)
np.save(
"./train_data/ddpg_enc_actions" + str(self.state_size) +
str(self.n_neighbors) + "_scores", scores)
np.save(
"./train_data/ddpg_enc_actions" + str(self.state_size) +
str(self.n_neighbors) + "_time", t_time)
np.save(
"./train_data/ddpg_enc_actions" + str(self.state_size) +
str(self.n_neighbors) + "_steps", steps)
def train_exploiting_greedy(self,
episodes=1000,
max_steps=1000,
plot_rewards=True):
scores = []
e = 0
for _ in range(episodes):
trace = []
greedy_reversal_sort(self.env.observation_space.sample(), trace)
for __ in range(3):
for permutation in trace[::-1]:
score = self.run_episode(max_steps, forced=permutation)
scores.append(score)
print("Episode:", e, " score:", score, " epsilon:",
self.epsilon)
e += 1
print()
print()
self.model.save_weights(FINAL_WEIGHTS_PATH)
scores = np.array(scores)
if plot_rewards:
plot(scores)
plot_running_avg(scores)
def train(self, episodes=1000, max_steps=1000, plot_rewards=True):
# Initialize target network weights
scores, steps = np.empty(episodes), np.empty(episodes)
start = time.time()
break_flag = 0
for e in range(episodes):
score, step, loss = self.run_episode(max_steps)
scores[e], steps[e] = score, step
print("Episode:", e, " steps:", step, " score:", score,
" loss:", loss, " time:",
time.time() - start)
#break_flag = break_flag+1 if step == max_steps else 0
#if break_flag > 60: break
saver = tf.train.Saver()
saver.save(self.session, self.train_path)
if plot_rewards:
t_time = time.time() - start
print("Mean step:", np.mean(steps), " Total steps:", np.sum(steps),
" total time:", t_time)
np.save(
"./train_data/ddpg_enc_actions" + str(self.state_size) + '_' +
str(self.n_neighbors) + "_scores", scores)
np.save(
"./train_data/ddpg_enc_actions" + str(self.state_size) + '_' +
str(self.n_neighbors) + "_time", t_time)
np.save(
"./train_data/ddpg_enc_actions" + str(self.state_size) + '_' +
str(self.n_neighbors) + "_steps", steps)
plot(steps)
plot_running_avg(steps)
def train(self, episodes=1000, max_steps=1000, plot_rewards=True):
scores, steps = np.empty(episodes), np.empty(episodes)
start = time.time()
for e in range(episodes):
score, step = self.run_episode(max_steps)
scores[e], steps[e] = score, step
print("Episode:", e, " steps:", step, " score:", score,
" epsilon:", self.epsilon, " time:",
time.time() - start)
'''if e%100 == 0:
ensure_saved_models_dir()
self.model.save_weights(FINAL_WEIGHTS_PATH)
print("Weights Saved")'''
ensure_saved_models_dir()
self.model.save_weights(FINAL_WEIGHTS_PATH)
if plot_rewards:
t_time = time.time() - start
print("Mean score:", np.mean(scores), " Total steps:",
np.sum(steps), " total time:", t_time)
plot(scores)
plot_running_avg(scores)
np.save("./train_data/ddqn_" + str(self.state_size) + "_scores",
scores)
np.save("./train_data/ddqn_" + str(self.state_size) + "_time",
t_time)
np.save("./train_data/ddqn_" + str(self.state_size) + "_steps",
steps)
def train(self, episodes=1000, max_steps=1000, plot_rewards=True):
scores = np.empty(episodes)
for e in range(episodes):
score = self.run_episode(max_steps)
scores[e] = score
print("Episode:", e, " score:", score, " epsilon:", self.epsilon)
ensure_saved_models_dir()
self.model.save_weights(FINAL_WEIGHTS_PATH)
if plot_rewards:
plot(scores)
plot_running_avg(scores)
def train(self, n, f_eps, plot_rewards=False, plot_best=False):
total_rewards = np.empty(n)
bests = np.empty(n)
best = None
for i in range(n):
eps = f_eps(i)
total_reward = self._play_one(eps)
if best is None or total_reward > best:
best = total_reward
bests[i] = best
total_rewards[i] = total_reward
if i % 1 == 0:
print("Episode:", i, "Reward:", total_reward, "Best:", best,
"Eps:", eps)
if plot_rewards:
plot_running_avg(total_rewards)
if plot_best:
plot(bests, 'Bests')
def train(self, episodes=1000, max_steps=800, plot_rewards=True):
scores, steps, losses = np.zeros(episodes), np.zeros(episodes), np.zeros(episodes)
start = time.time()
saver = tf.train.Saver()
if self.fill_mem:
self.fill_memory()
for e in range(episodes):
score, step, loss = self.run_episode(max_steps)
scores[e], steps[e], losses[e] = score, step, loss
print("Episode:", e, " steps:", step, " score: %.1f" % score," loss:", loss, " epsilon:", self.epsilon, " time:", time.time() - start)
if math.isnan(loss): break
ensure_saved_models_dir()
saver.save(self.session, self.train_path)
if plot_rewards:
t_time = time.time() - start
print("Mean score:", np.mean(scores), " Total steps:", np.sum(steps), " total time:", t_time)
np.save("./train_data/ddqn_tf_" + str(self.state_size) + "_scores", scores)
np.save("./train_data/ddqn_tf_" + str(self.state_size) + "_time", t_time)
np.save("./train_data/ddqn_tf_" + str(self.state_size) + "_steps", steps)
plot(steps)
plot_running_avg(steps)
plot_running_avg(losses, title="Losses")
def train(self, episodes=1000, max_steps=800, plot_rewards=True):
# Initialize target network weights
self.actor.update_target_model(copy=True)
self.critic.update_target_model(copy=True)
scores, steps = np.empty(episodes), np.empty(episodes)
start = time.time()
break_flag = 0
for e in range(episodes):
score, step = self.run_episode(max_steps)
scores[e], steps[e] = score, step
print("Episode:", e, " steps:", step, " score:", score,
" time:",
time.time() - start)
if e % 50 == 0 and step == max_steps and self.fill_mem:
self.fill_memory()
break_flag = break_flag + 1 if step == max_steps else 0
if break_flag > 50 and e >= episodes / 2: break
ensure_saved_models_dir()
saver = tf.train.Saver()
saver.save(self.session, self.train_path)
if plot_rewards:
t_time = time.time() - start
print("Mean score:", np.mean(scores), " Total steps:",
np.sum(steps), " total time:", t_time)
np.save(
"./train_data/ddpg_enc_actions" + str(self.state_size) +
str(self.n_neighbors) + "_scores", scores)
np.save(
"./train_data/ddpg_enc_actions" + str(self.state_size) +
str(self.n_neighbors) + "_time", t_time)
np.save(
"./train_data/ddpg_enc_actions" + str(self.state_size) +
str(self.n_neighbors) + "_steps", steps)
plot(steps)
plot_running_avg(steps)