tf.reduce.mean()用法

感谢：https://tensorflow.google.cn/api_docs/python/tf/math/reduce_mean

别名:

• tf.math.reduce_mean

• tf.reduce_mean 两种叫法都可以~

  tf.math.reduce_mean(
      input_tensor,
      axis=None,
      keepdims=None,
      name=None,
      reduction_indices=None,
      keep_dims=None
  )

•  

Computes the mean of elements across dimensions of a tensor. (deprecated arguments)

计算Tensor元素的平均值的吧

一些ARGUMENTS已经被弃用，在未来版本将被删除。keepdims将代替keep_dims

Reduces input_tensor along the dimensions given in axis. Unless keepdims is true, the rank of the tensor is reduced by 1 for each entry in axis. If keepdims is true, the reduced dimensions are retained with length 1.

If axis is None, all dimensions are reduced, and a tensor with a single element is returned.

当axis=None的时候，计算全部维度的平均值，当axis=0时，计算每列的平均值，当axis=1时，计算每行的平均值

For example:

x = tf.constant([[1., 1.], [2., 2.]])
tf.reduce_mean(x)  # 1.5
tf.reduce_mean(x, 0)  # [1.5, 1.5]
tf.reduce_mean(x, 1)  # [1.,  2.]

Args:

• input_tensor: The tensor to reduce. Should have numeric type.
• axis: The dimensions to reduce. If None (the default), reduces all dimensions. Must be in the range[-rank(input_tensor), rank(input_tensor)).
• keepdims: If true, retains reduced dimensions with length 1.
• name: A name for the operation (optional).
• reduction_indices: The old (deprecated) name for axis.
• keep_dims: Deprecated alias for keepdims.

Returns:

The reduced tensor.

Numpy Compatibility

Equivalent to np.mean

Please note that np.mean has a dtype parameter that could be used to specify the output type. By default this is dtype=float64. On the other hand, tf.reduce_mean has an aggressive type inference from input_tensor, for example:

x = tf.constant([1, 0, 1, 0])
tf.reduce_mean(x)  # 0
y = tf.constant([1., 0., 1., 0.])
tf.reduce_mean(y)  # 0.5