利用了三個 實例，中間一直出問題。Residual-Dense-Network-Trained-with-cGAN-for-Super-Resolution-master、Spectral_Normalization-Tensorflow-master、SNGAN-master。最終使用了 第一個中的方法。
TripleGAN-Tensorflow-SN中。
具體代碼如下：黑體爲修改的代碼：
1、ops.py中
import tensorflow as tf
#import tflearn
from tflearn import global_avg_pool
from tensorflow.contrib.layers import variance_scaling_initializer
import numpy as np
import math
he_init = variance_scaling_initializer()
# he_init = tf.truncated_normal_initializer(stddev=0.02)
"""
The weight norm is not implemented at this time.
"""
def weight_norm(x, output_dim) :
input_dim = int(x.get_shape()[-1])
g = tf.get_variable('g_scalar', shape=[output_dim], dtype=tf.float32, initializer=tf.ones_initializer())
w = tf.get_variable('weight', shape=[input_dim, output_dim], dtype=tf.float32, initializer=he_init)
w_init = tf.nn.l2_normalize(w, dim=0) * g # SAME dim=1
return tf.variables_initializer(w_init)
weight_init = tf.random_normal_initializer(mean=0.0, stddev=0.02)
weight_regularizer = None
def conv_layer(x, filter_size, kernel, stride=1, padding='SAME', wn=False,sn=True, layer_name="conv"):
with tf.name_scope(layer_name):
'''if wn:
w_init = weight_norm(x, filter_size)
x = tf.layers.conv2d(inputs=x, filters=filter_size, kernel_size=kernel, kernel_initializer=w_init, strides=stride, padding=padding)
else :
x = tf.layers.conv2d(inputs=x, filters=filter_size, kernel_size=kernel, kernel_initializer=he_init, strides=stride, padding=padding)
'''
if sn :
with tf.variable_scope('scope', reuse=tf.AUTO_REUSE):
w = tf.get_variable("kernel"+layer_name, shape=[kernel[0], kernel[1], x.get_shape()[-1], filter_size], initializer=weight_init,regularizer=weight_regularizer)
print("w",w)
#bias = tf.get_variable("bias", [filter_size], initializer=tf.constant_initializer(0.0))
x = tf.nn.conv2d(input=x, filter=spectral_norm(layer_name,w), strides=[1, stride, stride, 1], padding=padding)
print("x",x)
else :
x = tf.layers.conv2d(inputs=x, filters=filter_size,
kernel_size=kernel, kernel_initializer=weight_init,
kernel_regularizer=weight_regularizer,
strides=stride,padding=padding)
return x
def deconv_layer(x, filter_size, kernel, stride=1, padding='SAME', wn=False, sn=False, layer_name='deconv'):
with tf.name_scope(layer_name):
if wn :
w_init = weight_norm(x, filter_size)
x = tf.layers.conv2d_transpose(inputs=x, filters=filter_size, kernel_size=kernel, kernel_initializer=w_init, strides=stride, padding=padding)
else :
x = tf.layers.conv2d_transpose(inputs=x, filters=filter_size, kernel_size=kernel, kernel_initializer=he_init, strides=stride, padding=padding)
x_shape = x.get_shape().as_list()
output_shape = [x_shape[0], x_shape[1] * stride, x_shape[2] * stride, filter_size]
if sn:
w = tf.get_variable("kernel", shape=[kernel[0], kernel[1], filter_size, x.get_shape()[-1]], initializer=weight_init,
regularizer=weight_regularizer)
x = tf.nn.conv2d_transpose(x, filter=spectral_normed_weight(w), output_shape=output_shape,
strides=[1, stride, stride, 1], padding='SAME')
else:
x = tf.layers.conv2d_transpose(inputs=x, filters=filter_size,
kernel_size=kernel, kernel_initializer=weight_init,
kernel_regularizer=weight_regularizer,
strides=stride, padding='SAME')
return x
def linear(x, unit, wn=False, layer_name='linear'):
with tf.name_scope(layer_name):
if wn :
w_init = weight_norm(x, unit)
x = tf.layers.dense(inputs=x, units=unit, kernel_initializer=w_init)
else :
x = tf.layers.dense(inputs=x, units=unit, kernel_initializer=he_init)
return x
def nin(x, unit, wn=False, layer_name='nin'):
# https://github.com/openai/weightnorm/blob/master/tensorflow/nn.py
with tf.name_scope(layer_name):
s = list(map(int, x.get_shape()))
x = tf.reshape(x, [np.prod(s[:-1]), s[-1]])
x = linear(x, unit, wn, layer_name)
x = tf.reshape(x, s[:-1] + [unit])
return x
def gaussian_noise_layer(x, std=0.15):
noise = tf.random_normal(shape=tf.shape(x), mean=0.0, stddev=std, dtype=tf.float32)
return x + noise
def Global_Average_Pooling(x):
return global_avg_pool(x, name='Global_avg_pooling')
def max_pooling(x, kernel, stride):
return tf.layers.max_pooling2d(x, pool_size=kernel, strides=stride, padding='VALID')
def flatten(x):
return tf.contrib.layers.flatten(x)
def lrelu(x, leak=0.2, name="lrelu"):
return tf.maximum(x, leak * x)
def sigmoid(x):
return tf.nn.sigmoid(x)
def relu(x):
return tf.nn.relu(x)
def tanh(x):
return tf.nn.tanh(x)
def conv_concat(x, y):
x_shapes = x.get_shape()
y_shapes = y.get_shape()
return concat([x, y * tf.ones([x_shapes[0], x_shapes[1], x_shapes[2], y_shapes[3]])], axis=3)
def concat(x, axis=1):
return tf.concat(x, axis=axis)
def reshape(x, shape):
return tf.reshape(x, shape=shape)
def batch_norm(x, is_training, scope):
return tf.contrib.layers.batch_norm(x,
decay=0.9,
updates_collections=None,
epsilon=1e-5,
scale=True,
is_training=is_training,
scope=scope)
def instance_norm(x, is_training, scope):
with tf.variable_scope(scope):
epsilon = 1e-5
mean, var = tf.nn.moments(x, [1, 2], keep_dims=True)
scale = tf.get_variable('scale', [x.get_shape()[-1]],
initializer=tf.truncated_normal_initializer(mean=1.0, stddev=0.02))
offset = tf.get_variable('offset', [x.get_shape()[-1]], initializer=tf.constant_initializer(0.0))
out = scale * tf.div(x - mean, tf.sqrt(var + epsilon)) + offset
return out
def dropout(x, rate, is_training):
return tf.layers.dropout(inputs=x, rate=rate, training=is_training)
def rampup(epoch):
if epoch < 80:
p = max(0.0, float(epoch)) / float(80)
p = 1.0 - p
return math.exp(-p*p*5.0)
else:
return 1.0
def rampdown(epoch):
if epoch >= (300 - 50):
ep = (epoch - (300 - 50)) * 0.5
return math.exp(-(ep * ep) / 50)
else:
return 1.0
def _l2normalize(v, eps=1e-12):
return v / tf.sqrt(tf.reduce_sum(tf.square(v)) + eps)
def max_singular_value(W, u, Ip=1):
_u = u
_v = 0
for _ in range(Ip):
_v = _l2normalize(tf.matmul(_u, W), eps=1e-12)
_u = _l2normalize(tf.matmul(_v, W, transpose_b=True), eps=1e-12)
_v = tf.stop_gradient(_v)
_u = tf.stop_gradient(_u)
sigma = tf.reduce_sum(tf.matmul(_u, W) * _v)
return sigma, _u, _v
def spectral_norm(name, W, Ip=1):
u = tf.get_variable(name + "_u", [1, W.shape[-1]], initializer=tf.random_normal_initializer(), trainable=False) # 1 x ch
W_mat = tf.transpose(tf.reshape(W, [-1, W.shape[-1]]))
sigma, _u, _ = max_singular_value(W_mat, u, Ip)
with tf.control_dependencies([tf.assign(u, _u)]):
W_sn = W / sigma
return W_sn
下面是一些問題：
Any ideas how can I solve problem shown below? With the information that I found on the web it is associated with problem of reusing tensorflow scope however nothing works.
ValueError: Variable rnn/basic_rnn_cell/kernel already exists, disallowed. Did you mean to set reuse=True or reuse=tf.AUTO_REUSE in VarScope? Originally defined at:
File "/code/backend/management/commands/RNN.py", line 370, in predict
states_series, current_state = tf.nn.dynamic_rnn(cell=cell, inputs=batchX_placeholder, dtype=tf.float32)
File "/code/backend/management/commands/RNN.py", line 499, in Command
predict("string")
File "/code/backend/management/commands/RNN.py", line 12, in <module>
class Command(BaseCommand):
I tried for instance something like this
with tf.variable_scope('scope'):
states_series, current_state = tf.nn.dynamic_rnn(cell=cell, inputs=batchX_placeholder, dtype=tf.float32)
and this
with tf.variable_scope('scope', reuse = True ):
states_series, current_state = tf.nn.dynamic_rnn(cell=cell, inputs=batchX_placeholder, dtype=tf.float32)
and this
with tf.variable_scope('scope', reuse = tf.AUTO_REUSE ):
states_series, current_state = tf.nn.dynamic_rnn(cell=cell, inputs=batchX_placeholder, dtype=tf.float32)
Any ideas?
python python-3.x machine-learning tensorflow neural-network
asked Nov 14 '17 at 23:26
user7304253
1 Answer
53
Does this happen when you run the model for the first time (upon opening a new python console)?
If not, you need to clear you computational graph. You can do that by putting this line at the beginning of your script.
tf.reset_default_graph()
-
Thank you Sir! I couldn't find any solution for some time, so I had to restart Spyder environment every time I wanted to run my NN! You saved me! – Ladislav Ondris Apr 10 '18 at 15:16
-
1
Thank you, restarting kernel and reloading dataset was so annoying until I found this. – Michal Fašánek Apr 14 '18 at 22:53
Tensorflow報錯：ValueError: Trying to share variable CON/conv2/W, but specified shape (3, 3, 128, 256) and found shape (3, 3, 128, 128).
我的使用情景是這樣的：
我在一個卷積層提供了一個filter，並且它的shape爲(3, 3, 128, 256)， 然後我通過tf.variable_scope等操作使這個filter的name爲CON/conv2/W。在使用tf.get_variable生成初始變量時就報了這樣一個bug。
部分代碼流程：
...
conv3 = conv_layer_new(relu2, [3, 3, 128, 256], strides = [1, 2, 2, 1], name = 'conv2')
...
def conv_layer_new(input, filter_shape, strides, name = None):
with tf.variable_scope(name):
W = tf.get_variable("W", filter_shape, initializer = tf. truncated_normal_initializer(0., 0.005))
...
重點來了：
那爲什麼它會自動給我一個(3, 3, 128, 128)這樣一個特定的shape呢？因爲報錯的原話是 found !
所以我就思考難道是之前有相同name的filter，它的shape就是(3, 3, 128, 128)？
結果證明我是對的，確實在這個出錯的卷積層使用了與之前相的name，導致調用了之前的shape。我將conv2改爲了conv3之後就正常了。（也就是name變爲了CON/conv3/W）
所以說博主寫了這麼多廢話就是想告訴大家一定要檢查一下前面的變量名，是否寫重複了，因爲如果有相同的name，tensorflow就會自作主張地進行調用！（博主比較傻，這個問題卡了我好久所以寫篇博客怕自己下次再犯。。。）
---------------------
作者：yyhhlancelot
來源：CSDN
原文：https://blog.csdn.net/yyhhlancelot/article/details/82979235
版權聲明：本文爲博主原創文章，轉載請附上博文鏈接！