op_holder.append(tfVars[idx+total_vars//2].assign((var.value()*tau) + ((1-tau)*tfVars[idx+total_vars//2].value())))
return op_holder
def updateTarget(op_holder,sess):
for op in op_holder:
sess.run(op)
def saveScore(score):
my_file = open(reward_savefile, 'a') # Name and path of the reward text file
my_file.write("%s\n" % score)
my_file.close()
###########################################
game = GameSimulator()
game.initialize()
ACTION_COUNT = game.get_action_size()
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0.33)
SESSION = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
if LOAD_MODEL:
EPSILON_MAX = 0.25 # restart after 20+ epoch
agent = Agent(memory_cap = MEMORY_CAP, batch_size = BATCH_SIZE, resolution = RESOLUTION, action_count = ACTION_COUNT,
session = SESSION, lr = LEARNING_RATE, gamma = GAMMA, epsilon_min = EPSILON_MIN, trace_length=TRACE_LENGTH,
epsilon_decay_steps = EPSILON_DECAY_STEPS, epsilon_max=EPSILON_MAX, hidden_size=HIDDEN_SIZE)
saver = tf.train.Saver()
