我们从Python开源项目中，提取了以下12个代码示例，用于说明如何使用chainer.optimizers.RMSpropGraves()。
def __init__(self,gpu,batchsize,data_dir,dataset,net,mode,epochs,save_every,size,**kwargs): super(Network, self).__init__(epochs,save_every) print "building ..." self.input_height=size self.input_width=size self.net = net self.mode=mode self.dataset=dataset self.train_data, self.test_data=self.get_dataset(data_dir,dataset) print 'input_channel ==> %d using %s dataset'%(self.in_channel, self.dataset) self.enc = GoogLeNet() self.dec = Decoder(self.in_size) self.xp = cuda.cupy cuda.get_device(gpu).use() self.enc.to_gpu() self.dec.to_gpu() self.o_dec = optimizers.RMSpropGraves() self.o_dec.setup(self.dec) self.batchsize=batchsize
def setup_optimizer(self): optimizer = optimizers.RMSpropGraves( lr=self.args.start_lr, alpha=0.95, momentum=0.9, eps=1e-08) optimizer.setup(self) optimizer.add_hook(chainer.optimizer.GradientClipping(self.args.grad_clip)) optimizer.add_hook(chainer.optimizer.WeightDecay(self.args.weight_decay)) return optimizer
def create(self): return optimizers.RMSpropGraves(0.1)
def reset_optimizer(self): self.optimizer = optimizers.RMSpropGraves(lr=0.00025, alpha=0.95, momentum=0.95, eps=0.0001) self.optimizer.setup(self.model)
def __init__(self, gpu_id, state_dimention,batchsize,historysize, enable_controller,arch): self.gpu_id = gpu_id self.num_of_actions = len(enable_controller) self.enable_controller = enable_controller # Default setting : "Pong" self.arch = arch self.replay_size = batchsize self.data_size = historysize self.state_dimention = state_dimention print "Initializing DQN..." # Initialization of Chainer 1.1.0 or older. # print "CUDA init" # cuda.init() print "Model Building" #self.model = dnn_6_new.Q_DNN(self.state_dimention,200,self.num_of_actions) self.model = self.set_model(self.arch,self.state_dimention,200,self.num_of_actions) self.model.to_gpu(self.gpu_id) self.model_target = copy.deepcopy(self.model) print "Initizlizing Optimizer" self.optimizer = optimizers.RMSpropGraves(lr=0.00025, alpha=0.95, momentum=0.95, eps=0.0001) self.optimizer.setup(self.model) # History Data : D=[s, a, r, s_dash, end_episode_flag] self.D = [np.zeros((self.data_size, 1, self.state_dimention), dtype=np.float32), np.zeros(self.data_size, dtype=np.int8), np.zeros((self.data_size, 1), dtype=np.float32), np.zeros((self.data_size, 1, self.state_dimention), dtype=np.float32), np.zeros((self.data_size, 1), dtype=np.bool)]
def __init__(self, gpu_id, state_dimention,batchsize,historysize, enable_controller=[1, -1, 0],targetFlag=False): self.gpu_id = gpu_id self.num_of_actions = len(enable_controller) self.enable_controller = enable_controller # Default setting : "Pong" self.replay_size = batchsize self.data_size = historysize self.state_dimention = state_dimention self.targetFlag = targetFlag print "Initializing DQN..." # Initialization of Chainer 1.1.0 or older. # print "CUDA init" # cuda.init() print "Model Building" self.model = dnn_6.Q_DNN(self.state_dimention,200,self.num_of_actions) self.model.to_gpu(self.gpu_id) self.model_target = copy.deepcopy(self.model) print "Initizlizing Optimizer" self.optimizer = optimizers.RMSpropGraves(lr=0.00025, alpha=0.95, momentum=0.95, eps=0.0001) self.optimizer.setup(self.model) # History Data : D=[s, a, r, s_dash, end_episode_flag] self.D = [np.zeros((self.data_size, 1, self.state_dimention), dtype=np.float32), np.zeros(self.data_size, dtype=np.int8), np.zeros((self.data_size, 1), dtype=np.float32), np.zeros((self.data_size, 1, self.state_dimention), dtype=np.float32), np.zeros((self.data_size, 1), dtype=np.bool)]
def __init__(self, use_gpu, enable_controller, dim): self.use_gpu = use_gpu self.num_of_actions = len(enable_controller) self.enable_controller = enable_controller self.dim = dim print("Initializing Q-Network...") hidden_dim = 256 self.model = FunctionSet( l4=F.Linear(self.dim*self.hist_size, hidden_dim, wscale=np.sqrt(2)), q_value=F.Linear(hidden_dim, self.num_of_actions, initialW=np.zeros((self.num_of_actions, hidden_dim), dtype=np.float32)) ) if self.use_gpu >= 0: self.model.to_gpu() self.model_target = copy.deepcopy(self.model) self.optimizer = optimizers.RMSpropGraves(lr=0.00025, alpha=0.95, momentum=0.95, eps=0.0001) self.optimizer.setup(self.model.collect_parameters()) # History Data : D=[s, a, r, s_dash, end_episode_flag] self.d = [np.zeros((self.data_size, self.hist_size, self.dim), dtype=np.uint8), np.zeros(self.data_size, dtype=np.uint8), np.zeros((self.data_size, 1), dtype=np.int8), np.zeros((self.data_size, self.hist_size, self.dim), dtype=np.uint8), np.zeros((self.data_size, 1), dtype=np.bool)]
def setup_optimizer(self): optimizer = optimizers.RMSpropGraves( lr=self.args.start_lr, alpha=0.95, momentum=0.9, eps=1e-08) optimizer.setup(self) optimizer.add_hook(chainer.optimizer.GradientClipping(self.args.grad_clip)) return optimizer
def __init__(self, state_shape, action_num, image_num_per_state, model, gamma=0.99, # discount factor replay_batch_size=32, replay_memory_size=5*10**4, target_model_update_freq=1, max_step=50, lr=0.00025, clipping=False # if True, ignore reward intensity ): print("initializing DQN...") self.action_num = action_num self.image_num_per_state = image_num_per_state self.gamma = gamma self.replay_batch_size = replay_batch_size self.replay_memory_size = replay_memory_size self.target_model_update_freq = target_model_update_freq self.max_step = max_step self.clipping = clipping print("Initializing Model...") self.model = model self.model_target = copy.deepcopy(self.model) print("Initializing Optimizer") self.optimizer = optimizers.RMSpropGraves(lr=lr, alpha=0.95, momentum=0.0, eps=0.01) self.optimizer.setup(self.model) self.optimizer.add_hook(chainer.optimizer.GradientClipping(20)) print("Initializing Replay Buffer...") self.dataset = dataset.DataSet( max_size=replay_memory_size, max_step=max_step, frame_shape=state_shape, frame_dtype=np.uint8) self.xp = model.xp self.state_shape = state_shape
def optimize_rmsprop(self, init_img, lr=0.1, alpha=0.95, momentum=0.9, eps=1e-4, iterations=2000, save=50, filename='iter', str_contrast=False): chainer_rms = optimizers.RMSpropGraves(lr=lr, alpha=alpha, momentum=momentum, eps=eps) state = {'n': xp.zeros_like(init_img.data), 'g': xp.zeros_like(init_img.data), 'delta': xp.zeros_like(init_img.data)} out_img = Variable(xp.zeros_like(init_img.data), volatile=True) time_start = time.time() for epoch in range(iterations): loss = self.loss_total(init_img) loss.backward() loss.unchain_backward() # normalize gradient grad_l1_norm = xp.sum(xp.absolute(init_img.grad * init_img.grad)) init_img.grad /= grad_l1_norm if gpu_flag: chainer_rms.update_one_gpu(init_img, state) else: chainer_rms.update_one_cpu(init_img, state) init_img.zerograd() # save image every 'save' iteration if save != 0 and (epoch + 1) % save == 0: if self.preserve_color: init_img_lum = separate_lum_chr(init_img)[0] if gpu_flag: init_img_lum.to_gpu() out_img.copydata(init_img_lum + self.content_img_chr) else: out_img.copydata(init_img) save_image(out_img, filename + '_' + str(epoch + 1) + '.png', contrast=str_contrast) print("Image Saved at Iteration %.0f, Time Used: %.4f, Total Loss: %.4f" % ((epoch + 1), (time.time() - time_start), loss.data)) # rescale an array to [0, 255]
def __init__(self, agent, memory_size=10**4, replay_size=32, gamma=0.99, initial_exploration=10**4, target_update_freq=10**4, learning_rate=0.00025, epsilon_decay=1e-6, minimum_epsilon=0.1): self.agent = agent self.target = Q(self.agent.q.n_history, self.agent.q.n_action, on_gpu=self.agent.q.on_gpu) self.memory_size = memory_size self.replay_size = replay_size self.gamma = gamma self.initial_exploration = initial_exploration self.target_update_freq = target_update_freq self.learning_rate = learning_rate self.epsilon_decay = epsilon_decay self.minimum_epsilon = minimum_epsilon self._step = 0 # prepare memory for replay n_hist = self.agent.q.n_history size = self.agent.q.SIZE self.memory = [ np.zeros((memory_size, n_hist, size, size), dtype=np.float32), np.zeros(memory_size, dtype=np.uint8), np.zeros((memory_size, 1), dtype=np.float32), np.zeros((memory_size, n_hist, size, size), dtype=np.float32), np.zeros((memory_size, 1), dtype=np.bool) ] self.memory_text = [ "state", "action", "reward", "next_state", "episode_end" ] # prepare optimizer self.optimizer = optimizers.RMSpropGraves(lr=learning_rate, alpha=0.95, momentum=0.95, eps=0.01) self.optimizer.setup(self.agent.q) self._loss = 9 self._qv = 0
def __init__(self, enable_controller=[0, 1, 3, 4]): self.num_of_actions = len(enable_controller) self.enable_controller = enable_controller # Default setting : "Breakout" print "Initializing DDQN..." # Initialization of Chainer 1.1.0 or older. # print "CUDA init" # cuda.init() print "Model Building" self.model = FunctionSet( l1=F.Convolution2D(4, 32, ksize=8, stride=4, nobias=False, wscale=np.sqrt(2)), l2=F.Convolution2D(32, 64, ksize=4, stride=2, nobias=False, wscale=np.sqrt(2)), l3=F.Convolution2D(64, 64, ksize=3, stride=1, nobias=False, wscale=np.sqrt(2)), l4=F.Linear(3136, 512, wscale=np.sqrt(2)), q_value=F.Linear(512, self.num_of_actions, initialW=np.zeros((self.num_of_actions, 512), dtype=np.float32)) ).to_gpu() if args.resumemodel: # load saved model serializers.load_npz(args.resumemodel, self.model) print "load model from resume.model" self.model_target = copy.deepcopy(self.model) print "Initizlizing Optimizer" self.optimizer = optimizers.RMSpropGraves(lr=0.00025, alpha=0.95, momentum=0.95, eps=0.0001) self.optimizer.setup(self.model.collect_parameters()) # History Data : D=[s, a, r, s_dash, end_episode_flag] if args.resumeD1 and args.resumeD2: # load saved D1 and D2 npz_tmp1 = np.load(args.resumeD1) npz_tmp2 = np.load(args.resumeD2) self.D = [npz_tmp1['D0'], npz_tmp1['D1'], npz_tmp1['D2'], npz_tmp2['D3'], npz_tmp2['D4']] npz_tmp1.close() npz_tmp2.close() print "loaded stored D1 and D2" else: self.D = [np.zeros((self.data_size, 4, 84, 84), dtype=np.uint8), np.zeros(self.data_size, dtype=np.uint8), np.zeros((self.data_size, 1), dtype=np.int8), np.zeros((self.data_size, 4, 84, 84), dtype=np.uint8), np.zeros((self.data_size, 1), dtype=np.bool)] print "initialize D data"
