[英]Normalization of a tensor array
Experts. 专家。 I am new to DNN and Python.
我是DNN和Python的新手。 I am trying to use tensorflow to do some DNN learning work.
我正在尝试使用张量流来做一些DNN学习工作。 During my working, I came across a problem that I myself cannot solve.
在工作中,我遇到了一个我自己无法解决的问题。 In one step, I would like to normalize a tensor called "inputs".
第一步,我想将张量标准化为“输入”。 The normalization is simply take the maximum abs of a vector, and divide all the elements of the vector my the maximum abs.
归一化简单地取向量的最大abs,并将向量的所有元素除以最大abs。 But the following problem occured:
但是发生了以下问题:
ValueError Traceback (most recent call last) in () ()中的ValueError追溯(最近一次通话)
55 tmp_index = tf.argmax(tmp_abs,0)
56 tmp_index1 = tf.cast(tmp_index,dtype = tf.int32)
---> 57 inputs = inputs/tmp_abs[tmp_index1]
58
59 if index != len(Layers)-1:
InvalidArgumentError: Shape must be rank 1 but is rank 2 for 'hidden2_3/strided_slice' (op: 'StridedSlice') with input shapes: [?,1], [1,1], [1,1], [1]. InvalidArgumentError:对于输入形状为[?,1],[1,1],[1,1],[1]的'hidden2_3 / strided_slice'(op:'StridedSlice'),形状必须为等级1,但等级2。
Any advice will be appreciated. 任何建议将被认真考虑。 Thanks!
谢谢!
# input features and labels
x_ = tf.placeholder(name="input", shape=[None, 1], dtype=np.float32)
y_ = tf.placeholder(name="output", shape=[None, 1], dtype=np.float32)
# tf variables
Hidden = []
# Hidden Layers
for index, num_hidden in enumerate(Layers):
with tf.name_scope("hidden{}".format(index+1)):
if index == 0:
weights = tf.Variable(tf.truncated_normal([Fea_Size,num_hidden], stddev = get_stddev(Fea_Size,num_hidden)))
bias = tf.Variable(tf.zeros([num_hidden]))
else:
weights = tf.Variable(tf.truncated_normal([Layers[index-1], num_hidden], stddev = get_stddev(Layers[index-1], num_hidden)))
bias = tf.Variable(tf.zeros([num_hidden]))
inputs = x_ if index == 0 else Hidden[index-1]
if index !=0:
tmp_abs = tf.abs(inputs)
tmp_index = tf.argmax(tmp_abs,0)
tmp_index1 = tf.cast(tmp_index,dtype = tf.int32)
inputs = inputs/tmp_abs[tmp_index1]
if index != len(Layers)-1:
Hidden.append(tf.nn.relu(tf.matmul(inputs,weights) + bias))
else:
nonlin_model = tf.nn.relu(tf.matmul(inputs,weights) + bias)
nonlin_loss = tf.reduce_mean(tf.pow(nonlin_model - y_, 2), name='cost')
train_step_nonlin = tf.train.GradientDescentOptimizer(0.01).minimize(nonlin_loss)
The problem is with the fact that your inputs
has invariant shape for axis 0
. 问题在于您的
inputs
的axis 0
形状不变。
You can instead use: 您可以改用:
inputs = inputs/tf.reduce_max(tf.abs(inputs))
tf.abs
returns the absolute value of inputs
. tf.abs
返回inputs
的绝对值。 tf.reduce_max
returns the maximum value. tf.reduce_max
返回最大值。
Here's a code snippet that worked for me: 这是一个对我有用的代码片段:
inputs = tf.placeholder(shape=[None, 1], dtype = tf.float32)
inputs_normal = inputs/tf.reduce_max(tf.abs(inputs))
all_pos = sess.run(inputs_normal, feed_dict={inputs:[[1],[2],[3]]})
all_neg = sess.run(inputs_normal, feed_dict={inputs:[[-1],[-2],[-3]]})
pos_neg = sess.run(inputs_normal, feed_dict={inputs:[[1],[2],[-3]]})
Here are the values: 这些是值:
all_pos = array([[0.33333334],
[0.6666667 ],
[1. ]], dtype=float32)
all_neg = array([[-0.33333334],
[-0.6666667 ],
[-1. ]], dtype=float32)
pos_neg = array([[ 0.33333334],
[ 0.6666667 ],
[-1. ]], dtype=float32)
I've shown it for 2-D
tensors, but it should work for higher dimensional tensors as well. 我已经针对
2-D
张量显示了它,但是它也应该适用于高维张量。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.