TypeError: int() argument must be a string, a bytes-like object or a number, not 'Tensor'

Question

I'm trying to convert a the initializer form from tf.Variable to tf.get_variable for Cudnn_GRU but I keep getting this error. I have to convert because tensorflow does not allow initializing in loop/control-flow functions and only allow lambda initializers or through tf.get_variable

I have reduced the problem into the following minimal example:

import tensorflow as tf
e = tf.random_uniform_initializer(-0.1, 0.1)
i = tf.constant(0)
def func():
    gru_fw = tf.contrib.cudnn_rnn.CudnnGRU(num_layers=1, num_units=75, input_size=25)
    # original line: commented out and working if not under a control flow mechanism
    # param_fw = tf.Variable(tf.random_uniform([gru_fw.params_size()], -0.1, 0.1), validate_shape=False)
    # converted line
    param_fw = tf.get_variable("abcd", shape=[gru_fw.params_size()],initializer=e, validate_shape=False)
    return param_fw

def func2():
    ### repeat the same thing from func1
    pass

result = tf.cond(tf.equal(i, tf.constant(0)),func,func2)

The traceback is as follows:

Traceback (most recent call last):
    File "test_run_error.py", line 16, in <module>
        result = tf.cond(tf.equal(i, tf.constant(0)),func,func2)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/util/deprecation.py", line 316, in new_func
        return func(*args, **kwargs)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/ops/control_flow_ops.py", line 1855, in cond
        orig_res_t, res_t = context_t.BuildCondBranch(true_fn)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/ops/control_flow_ops.py", line 1725, in BuildCondBranch
        original_result = fn()
    File "test_run_error.py", line 9, in func
        param_fw = tf.get_variable("abcd", shape=[gru_fw.params_size()],initializer=e, validate_shape=False)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/ops/variable_scope.py", line 1203, in get_variable
        constraint=constraint)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/ops/variable_scope.py", line 1092, in get_variable
        constraint=constraint)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/ops/variable_scope.py", line 425, in get_variable
        constraint=constraint)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/ops/variable_scope.py", line 394, in _true_getter
        use_resource=use_resource, constraint=constraint)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/ops/variable_scope.py", line 730, in _get_single_variable
        shape = tensor_shape.as_shape(shape)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/framework/tensor_shape.py", line 849, in as_shape
        return TensorShape(shape)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/framework/tensor_shape.py", line 455, in __init__
        self._dims = [as_dimension(d) for d in dims_iter]
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/framework/tensor_shape.py", line 455, in <listcomp>
        self._dims = [as_dimension(d) for d in dims_iter]
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/framework/tensor_shape.py", line 397, in as_dimension
        return Dimension(value)
    File "/home/search/snetP/snet/lib/python3.5/site-packages/tensorflow/python/framework/tensor_shape.py", line 32, in __init__
        self._value = int(value)
TypeError: int() argument must be a string, a bytes-like object or a number, not 'Tensor'

Could you include the full traceback resulting from running this code? — glibdud
– glibdud, Commented Feb 12, 2018 at 15:14

Stephen · Accepted Answer · 2018-02-14 04:45:23Z

2

The problem seems to be that gru_fw.params_size() is returning Tensor("strided_slice_1:0", shape=(), dtype=int32) instead of the int it's apparently supposed to return. tf.get_variable won't accept a tensor as the shape argument. Your tf.Variable code runs but it produces a variable with a shape of <unknown>, which would likely cause problems when you try to use it.

Unfortunately I'm not finding a lot of documentation on how to properly create and use a CudnnGRU object. Are you following a tutorial from somewhere? Also, what version of TensorFlow are you using?

answered Feb 14, 2018 at 4:45

Stephen

8241 gold badge8 silver badges16 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

burglarhobbit Over a year ago

The tf.Variable code works perfectly as the code progresses further. I'm following a project. Tensorflow version is 1.4.1 github.com/HKUST-KnowComp/R-Net/blob/master/func.py

MEJBAH AHAMMAD · Accepted Answer · 2020-08-28 20:09:51Z

0

This method works for me, Try it out.

layers1= tf.compat.v1.layers.dense(tf_x, 10, tf.nn.relu)
output= tf.compat.v1.layers.dense(layers1, 1)

answered Aug 28, 2020 at 20:09

MEJBAH AHAMMAD

13 bronze badges

1 Comment

Joe Ferndz Over a year ago

Can you be more clear on where this code should be incorporated. Is this replacing an existing code or is this in addition to an existing code. Also, I would recommend you read how to write a good answer.

Collectives™ on Stack Overflow

TypeError: int() argument must be a string, a bytes-like object or a number, not 'Tensor'

2 Answers 2

1 Comment

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related