我曾经使用此代码来训练变分自编码器(我在论坛上找到了该代码并根据我的需要对其进行了调整):
import pickle
from pylab import mpl,plt
#lecture des résultats
filename=r'XXX.pic'
data_file=open(filename,'rb')
X_sec = pickle.load(data_file)#[:,3000:]
data_file.close()
size=X_sec.shape[0]
prop=0.75
cut=int(size*prop)
X_train=X_sec[:cut]
X_test=X_sec[cut:]
std=X_train.std()
X_train /= std
X_test /= std
import keras
from keras import layers
from keras import backend as K
from keras.models import Model
import numpy as np
#encoding_dim = 12
sig_shape = (3600,)
batch_size = 128
latent_dim = 12
input_sig = keras.Input(shape=sig_shape)
x = layers.Dense(128, activation='relu')(input_sig)
x = layers.Dense(64, activation='relu')(x)
shape_before_flattening = K.int_shape(x)
x = layers.Dense(32, activation='relu')(x)
z_mean = layers.Dense(latent_dim)(x)
z_log_var = layers.Dense(latent_dim)(x)
encoder=Model(input_sig,[z_mean,z_log_var])
def sampling(args):
z_mean, z_log_var = args
epsilon = K.random_normal(shape=(K.shape(z_mean)[0], latent_dim),
mean=0., stddev=1.)
return z_mean + K.exp(z_log_var) * epsilon
z = layers.Lambda(sampling)([z_mean, z_log_var])
decoder_input = layers.Input(K.int_shape(z)[1:])
x = layers.Dense(np.prod(shape_before_flattening[1:]),activation='relu')(decoder_input)
x = layers.Reshape(shape_before_flattening[1:])(x)
x = layers.Dense(128, activation='relu')(x)
x = layers.Dense(3600, activation='linear')(x)
decoder = Model(decoder_input, x)
z_decoded = decoder(z)
class CustomVariationalLayer(keras.layers.Layer):
def vae_loss(self, x, z_decoded):
x = K.flatten(x)
z_decoded = K.flatten(z_decoded)
xent_loss = keras.metrics.mae(x, z_decoded)
kl_loss = -5e-4 * K.mean(
1 + z_log_var - K.square(z_mean) - K.exp(z_log_var), axis=-1)
return K.mean(xent_loss + kl_loss)
def call(self, inputs):
x = inputs[0]
z_decoded = inputs[1]
loss = self.vae_loss(x, z_decoded)
self.add_loss(loss, inputs=inputs)
return x
y = CustomVariationalLayer()([input_sig, z_decoded])
vae = Model(input_sig, y)
vae.compile(optimizer='rmsprop', loss=None)
vae.summary()
vae.fit(x=X_train, y=None,shuffle=True,epochs=100,batch_size=batch_size,validation_data=(X_test, None))
它曾经运行顺利,但我已经更新了我的库,现在我收到了这个错误:
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\framework\ops.py”,第 1619 行,在 _create_c_op c_op = c_api.TF_FinishOperation(op_desc )
InvalidArgumentError:图中节点名称重复:“lambda_1/random_normal/shape”
在处理上述异常的过程中,又发生了一个异常:
回溯(最近一次调用最后一次):
文件“I:\Documents\Nico\Python\finance\travail_amont\autoencoder_variationnel_bruit.py”,第 74 行,在 z = layers.Lambda(sampling)([z_mean, z_log_var])
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\keras\backend\tensorflow_backend.py”,第75行,symbolic_fn_wrapper return func(*args, **kwargs )
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\keras\engine\base_layer.py”,第 506 行,调用output_shape = self.compute_output_shape(input_shape)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\keras\layers\core.py”,第674行,在compute_output_shape x = self.call(xs)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\keras\layers\core.py”,第 716 行,调用 return self.function(inputs, **论据)
文件“I:\Documents\Nico\Python\finance\travail_amont\autoencoder_variationnel_bruit.py”,第 71 行,采样均值=0,stddev=1。)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\keras\backend\tensorflow_backend.py”,第 4329 行,在 random_normal 形状中,mean=mean,stddev=stddev , dtype=dtype, 种子=种子)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\keras\backend.py”,第5602行,random_normal 形状,mean=mean,stddev =stddev,dtype=dtype,种子=种子)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\ops\random_ops.py”,第 69 行,在 random_normal shape_tensor = tensor_util.shape_tensor(shape )
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\framework\tensor_util.py”,第 994 行,shape_tensor 返回 ops.convert_to_tensor(shape, dtype=dtype,名称=“形状”)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\framework\ops.py”,第 1314 行,在 convert_to_tensor ret = conversion_func(value, dtype =dtype,名称=名称,as_ref=as_ref)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\ops\array_ops.py”,第 1368 行,在 _autopacking_conversion_function return _autopacking_helper(v, dtype,名称或“打包”)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\ops\array_ops.py”,第 1304 行,在 _autopacking_helper 中返回 gen_array_ops.pack(elems_as_tensors,名称=范围)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\ops\gen_array_ops.py”,第 5704 行,在包“Pack”中,值=值,轴=轴,名称=名称)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\framework\op_def_library.py”,第742行,_apply_op_helper attrs=attr_protos, op_def=op_def )
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\framework\func_graph.py”,第 595 行,在 _create_op_internal compute_device 中)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\framework\ops.py”,第 3322 行,在 _create_op_internal op_def=op_def 中)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\framework\ops.py”,第 1786 行,在init control_input_ops 中)
文件“C:\Users\user\AppData\Local\conda\conda\envs\my_root\lib\site-packages\tensorflow_core\python\framework\ops.py”,第 1622 行,在 _create_c_op 中引发 ValueError(str(e) )
ValueError:图中的重复节点名称:'lambda_1/random_normal/shape'
我不知道这个错误:“图中节点名称重复”。有没有人有线索?谢谢。
本文收集自互联网,转载请注明来源。
如有侵权,请联系 [email protected] 删除。
我来说两句