Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Tensorflow: How to get a tensor by name?

I'm having trouble recovering a tensor by name, I don't even know if it's possible.

I have a function that creates my graph:

def create_structure(tf, x, input_size,dropout):      with tf.variable_scope("scale_1") as scope:   W_S1_conv1 = deep_dive.weight_variable_scaling([7,7,3,64], name='W_S1_conv1')   b_S1_conv1 = deep_dive.bias_variable([64])   S1_conv1 = tf.nn.relu(deep_dive.conv2d(x_image, W_S1_conv1,strides=[1, 2, 2, 1], padding='SAME') + b_S1_conv1, name="Scale1_first_relu") . . . return S3_conv1,regularizer 

I want to access the variable S1_conv1 outside this function. I tried:

with tf.variable_scope('scale_1') as scope_conv:   tf.get_variable_scope().reuse_variables()  ft=tf.get_variable('Scale1_first_relu') 

But that is giving me an error:

ValueError: Under-sharing: Variable scale_1/Scale1_first_relu does not exist, disallowed. Did you mean to set reuse=None in VarScope?

But this works:

with tf.variable_scope('scale_1') as scope_conv:   tf.get_variable_scope().reuse_variables()  ft=tf.get_variable('W_S1_conv1') 

I can get around this with

return S3_conv1,regularizer, S1_conv1 

but I don't want to do that.

I think my problem is that S1_conv1 is not really a variable, it's just a tensor. Is there a way to do what I want?

like image 443
protas Avatar asked Apr 14 '16 01:04

protas


People also ask

How do you find the value of tensor in TensorFlow?

The easiest[A] way to evaluate the actual value of a Tensor object is to pass it to the Session. run() method, or call Tensor. eval() when you have a default session (i.e. in a with tf. Session(): block, or see below).

How do you identify a tensor type?

We can access the data type of a tensor using the ". dtype" attribute of the tensor. It returns the data type of the tensor.

Does TensorFlow use tensors?

Learn about Tensors, the multi-dimensional arrays used by TensorFlow. Tensors are multi-dimensional arrays with a uniform type (called a dtype ).


2 Answers

There is a function tf.Graph.get_tensor_by_name(). For instance:

import tensorflow as tf  c = tf.constant([[1.0, 2.0], [3.0, 4.0]]) d = tf.constant([[1.0, 1.0], [0.0, 1.0]]) e = tf.matmul(c, d, name='example')  with tf.Session() as sess:     test =  sess.run(e)     print e.name #example:0     test = tf.get_default_graph().get_tensor_by_name("example:0")     print test #Tensor("example:0", shape=(2, 2), dtype=float32) 
like image 178
apfalz Avatar answered Sep 19 '22 12:09

apfalz


All tensors have string names which you can see as follows

[tensor.name for tensor in tf.get_default_graph().as_graph_def().node] 

Once you know the name you can fetch the Tensor using <name>:0 (0 refers to endpoint which is somewhat redundant)

For instance if you do this

tf.constant(1)+tf.constant(2) 

You have the following Tensor names

[u'Const', u'Const_1', u'add'] 

So you can fetch output of addition as

sess.run('add:0') 

Note, this is part not part of public API. Automatically generated string tensor names are an implementation detail and may change.

like image 29
Yaroslav Bulatov Avatar answered Sep 20 '22 12:09

Yaroslav Bulatov