def train_agent():
plot_scores = []
plot_mean_scores = []
total_score = 0
record = 0
agent = Agent()
game = SnakeGame()
while True:
# get old state
state_old = agent.get_state(game)
# get move
final_move = agent.get_action(state_old)
# perform move and get new state
reward, done, score = game.play_step(final_move)
state_new = agent.get_state(game)
# train short memory
agent.train_short_memory(state_old, final_move, reward, state_new,
done)
# remember
agent.remember(state_old, final_move, reward, state_new, done)
if done:
game.reset()
agent.num_games += 1
agent.train_long_memory()
if score > record:
record = score
agent.model.save()
print(f'Game: {agent.num_games}, Score: {score}, Record: {record}')
plot_scores.append(score)
total_score += score
mean_score = total_score / agent.num_games
plot_mean_scores.append(mean_score)
plot_scores(plot_scores, plot_mean_scores)
if not model:
model = neural_network_model(input_size = len(X[0]))
model.fit({'input': X}, {'targets': y}, n_epoch=7, snapshot_step=500, show_metric=True, run_id='openai_learning')
return model
training_data = initial_population()
model = train_model(training_data)
scores = []
choices = []
for each_game in range(10):
score = 0
game_memory = []
game.reset( True )
win = game.getWindow()
alive = True
prev_obs = []
new_observation = []
while(True):
win.getch()
if len(prev_obs)==0:
action = game.sample()
else:
action = np.argmax(model.predict(prev_obs.reshape(-1,len(prev_obs),1))[0])
choices.append(action)
new_observation, reward, alive = game.step(action)
if not alive: