Python TensorFlow：张量沿轴的最大值

TensorFlow: Max of a tensor along an axis

提问by aphdstudent

My question is in two connected parts:

我的问题分为两个相互关联的部分：

1. How do I calculate the max along a certain axis of a tensor? For example, if I have

   x = tf.constant([[1,220,55],[4,3,-1]])
   

   I want something like

   x_max = tf.max(x, axis=1)
   print sess.run(x_max)
   
   output: [220,4]
   

   I know there is a tf.argmaxand a tf.maximum, but neither give the maximum value along an axis of a single tensor. For now I have a workaround:

   x_max = tf.slice(x, begin=[0,0], size=[-1,1])
   for a in range(1,2):
       x_max = tf.maximum(x_max , tf.slice(x, begin=[0,a], size=[-1,1]))
   

   But it looks less than optimal. Is there a better way to do this?

2. Given the indices of an argmaxof a tensor, how do I index into another tensor using those indices? Using the example of xabove, how do I do something like the following:

   ind_max = tf.argmax(x, dimension=1)    #output is [1,0]
   y = tf.constant([[1,2,3], [6,5,4])
   y_ = y[:, ind_max]                     #y_ should be [2,6]
   

   I know slicing, like the last line, does not exist in TensorFlow yet (#206).

   My question is: what is the best workaround for my specific case (maybe using other methods like gather, select, etc.)?

   Additional information: I know xand yare going to be two dimensional tensors only!

1. 如何计算张量沿某个轴的最大值？例如，如果我有

   x = tf.constant([[1,220,55],[4,3,-1]])
   

   我想要类似的东西

   x_max = tf.max(x, axis=1)
   print sess.run(x_max)
   
   output: [220,4]
   

   我知道有 atf.argmax和 a tf.maximum，但都没有给出沿单个张量轴的最大值。现在我有一个解决方法：

   x_max = tf.slice(x, begin=[0,0], size=[-1,1])
   for a in range(1,2):
       x_max = tf.maximum(x_max , tf.slice(x, begin=[0,a], size=[-1,1]))
   

   但它看起来不太理想。有一个更好的方法吗？

2. 给定argmax一个张量的索引，我如何使用这些索引索引另一个张量？使用x上面的示例，我如何执行以下操作：

   ind_max = tf.argmax(x, dimension=1)    #output is [1,0]
   y = tf.constant([[1,2,3], [6,5,4])
   y_ = y[:, ind_max]                     #y_ should be [2,6]
   

   我知道切片，就像最后一行一样，在 TensorFlow 中还不存在（#206）。

   我的问题是：对于我的特定情况（可能使用其他方法，如收集、选择等），最好的解决方法是什么？

   附加信息：我知道x并且y将只是二维张量！

采纳答案by mrry

The tf.reduce_max()operator provides exactly this functionality. By default it computes the global maximum of the given tensor, but you can specify a list of reduction_indices, which has the same meaning as axisin NumPy. To complete your example:

该tf.reduce_max()运营商提供的正是这种功能。默认情况下，它计算给定张量的全局最大值，但您可以指定 的列表reduction_indices，其含义与axisNumPy 中的含义相同。要完成您的示例：

x = tf.constant([[1, 220, 55], [4, 3, -1]])
x_max = tf.reduce_max(x, reduction_indices=[1])
print sess.run(x_max)  # ==> "array([220,   4], dtype=int32)"

If you compute the argmax using tf.argmax(), you could obtain the the values from a different tensor yby flattening yusing tf.reshape(), converting the argmax indices into vector indices as follows, and using tf.gather()to extract the appropriate values:

如果您使用 计算 argmax tf.argmax()，您可以y通过展平yusing tf.reshape()，将 argmax 索引转换为向量索引，并使用tf.gather()来提取适当的值，从而从不同的张量中获取值：

ind_max = tf.argmax(x, dimension=1)
y = tf.constant([[1, 2, 3], [6, 5, 4]])

flat_y = tf.reshape(y, [-1])  # Reshape to a vector.

# N.B. Handles 2-D case only.
flat_ind_max = ind_max + tf.cast(tf.range(tf.shape(y)[0]) * tf.shape(y)[1], tf.int64)

y_ = tf.gather(flat_y, flat_ind_max)

print sess.run(y_) # ==> "array([2, 6], dtype=int32)"

回答by kmario23

As of TensorFlow 1.10.0-dev20180626, tf.reduce_maxaccepts axisand keepdimskeyword arguments offering the similar functionality of numpy.max.

从TensorFlow 1.10.0- dev20180626 开始，tf.reduce_max接受axis和keepdims关键字参数提供与numpy.max.

In [55]: x = tf.constant([[1,220,55],[4,3,-1]])

In [56]: tf.reduce_max(x, axis=1).eval() 
Out[56]: array([220,   4], dtype=int32)

To have a resultant tensor of the same dimension as the input tensor, use keepdims=True

要获得与输入张量相同维度的合成张量，请使用 keepdims=True

In [57]: tf.reduce_max(x, axis=1, keepdims=True).eval()Out[57]: 
array([[220],
       [  4]], dtype=int32)

---

---

If the axisargument is not explicitly specified then the tensor level maximum element is returned (i.e. all axes are reduced).

如果axis未明确指定参数，则返回张量级最大元素（即所有轴都减少）。

In [58]: tf.reduce_max(x).eval()
Out[58]: 220