如何获取 Tensorflow 张量尺寸（形状）作为 int 值？

Question

假设我有一个 Tensorflow 张量。如何将张量的维度（形状）作为整数值？我知道有两种方法，tensor.get_shape() 和 tf.shape(tensor)，但我无法将形状值作为整数 int32 值。

例如，下面我创建了一个二维张量，我需要获取行数和列数为int32，这样我就可以调用reshape()来创建一个张量形状 (num_rows * num_cols, 1)。但是，方法 tensor.get_shape() returns 值作为 Dimension 类型，而不是 int32.

import tensorflow as tf
import numpy as np

sess = tf.Session()    
tensor = tf.convert_to_tensor(np.array([[1001,1002,1003],[3,4,5]]), dtype=tf.float32)

sess.run(tensor)    
# array([[ 1001.,  1002.,  1003.],
#        [    3.,     4.,     5.]], dtype=float32)

tensor_shape = tensor.get_shape()    
tensor_shape
# TensorShape([Dimension(2), Dimension(3)])    
print tensor_shape    
# (2, 3)

num_rows = tensor_shape[0] # ???
num_cols = tensor_shape[1] # ???

tensor2 = tf.reshape(tensor, (num_rows*num_cols, 1))    
# Traceback (most recent call last):
#   File "<stdin>", line 1, in <module>
#   File "/usr/local/lib/python2.7/site-packages/tensorflow/python/ops/gen_array_ops.py", line 1750, in reshape
#     name=name)
#   File "/usr/local/lib/python2.7/site-packages/tensorflow/python/framework/op_def_library.py", line 454, in apply_op
#     as_ref=input_arg.is_ref)
#   File "/usr/local/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 621, in convert_to_tensor
#     ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
#   File "/usr/local/lib/python2.7/site-packages/tensorflow/python/framework/constant_op.py", line 180, in _constant_tensor_conversion_function
#     return constant(v, dtype=dtype, name=name)
#   File "/usr/local/lib/python2.7/site-packages/tensorflow/python/framework/constant_op.py", line 163, in constant
#     tensor_util.make_tensor_proto(value, dtype=dtype, shape=shape))
#   File "/usr/local/lib/python2.7/site-packages/tensorflow/python/framework/tensor_util.py", line 353, in make_tensor_proto
#     _AssertCompatible(values, dtype)
#   File "/usr/local/lib/python2.7/site-packages/tensorflow/python/framework/tensor_util.py", line 290, in _AssertCompatible
#     (dtype.name, repr(mismatch), type(mismatch).__name__))
# TypeError: Expected int32, got Dimension(6) of type 'Dimension' instead.

Answer 1

要获取整数列表形式的形状，请执行 tensor.get_shape().as_list()。

要完成您的 tf.shape() 通话，请尝试 tensor2 = tf.reshape(tensor, tf.TensorShape([num_rows*num_cols, 1]))。或者你可以直接做 tensor2 = tf.reshape(tensor, tf.TensorShape([-1, 1])) 可以推断出它的第一个维度。

Answer 2

另一种解决方法是这样的：

tensor_shape[0].value

这将 return Dimension 对象的 int 值。

Answer 3

对于二维张量，您可以使用以下代码获取行数和列数作为 int32：

rows, columns = map(lambda i: i.value, tensor.get_shape())

Answer 4

在以后的版本中（使用 TensorFlow 1.14 测试）有一种更像 numpy 的方法来获取张量的形状。您可以使用 tensor.shape 来获取张量的形状。

tensor_shape = tensor.shape
print(tensor_shape)

Answer 5

2.0兼容答案：在Tensorflow 2.x (2.1)中，你可以得到张量的维度（形状）作为整数值，如下代码所示：

方法一（使用tf.shape）:

import tensorflow as tf
c = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
Shape = c.shape.as_list()
print(Shape)   # [2,3]

方法二（使用tf.get_shape()）:

import tensorflow as tf
c = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
Shape = c.get_shape().as_list()
print(Shape)   # [2,3]

Answer 6

另一个简单的解决方案是使用 map() 如下：

tensor_shape = map(int, my_tensor.shape)

这会将所有 Dimension 对象转换为 int

如何获取 Tensorflow 张量尺寸（形状）作为 int 值？

How to get Tensorflow tensor dimensions (shape) as int values?

python

artificial-intelligence

machine-learning

tensorflow