我添加了一个简单的补全函数，只使用numpy操作符:

   def probs_to_onehot(output_probabilities):
        argmax_indices_array = np.argmax(output_probabilities, axis=1)
        onehot_output_array = np.eye(np.unique(argmax_indices_array).shape[0])[argmax_indices_array.reshape(-1)]
        return onehot_output_array

它以一个概率矩阵作为输入:例如:

[[0.03038822 0.65810204 0.16549407 0.3797123] . [0.02771272 0.2760752 0.3280924 0.33458805]

它会返回

[[0 0 0 0]... [0 0 0 1]

创建一个有足够列的零数组b，即a.max() + 1。 然后，对于每一行i，设置第a[i]列为1。

>>> a = np.array([1, 0, 3])
>>> b = np.zeros((a.size, a.max() + 1))
>>> b[np.arange(a.size), a] = 1

>>> b
array([[ 0.,  1.,  0.,  0.],
       [ 1.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  1.]])

>>> values = [1, 0, 3]
>>> n_values = np.max(values) + 1
>>> np.eye(n_values)[values]
array([[ 0.,  1.,  0.,  0.],
       [ 1.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  1.]])

这是一个与维度无关的独立解决方案。

这将把任何非负整数的N维数组arr转换为一个N+1维数组one_hot，其中one_hot[i_1，…，i_N,c] = 1表示arr[i_1，…，i_N] = c.可以通过np恢复输入。argmax (one_hot, 1)

def expand_integer_grid(arr, n_classes):
    """

    :param arr: N dim array of size i_1, ..., i_N
    :param n_classes: C
    :returns: one-hot N+1 dim array of size i_1, ..., i_N, C
    :rtype: ndarray

    """
    one_hot = np.zeros(arr.shape + (n_classes,))
    axes_ranges = [range(arr.shape[i]) for i in range(arr.ndim)]
    flat_grids = [_.ravel() for _ in np.meshgrid(*axes_ranges, indexing='ij')]
    one_hot[flat_grids + [arr.ravel()]] = 1
    assert((one_hot.sum(-1) == 1).all())
    assert(np.allclose(np.argmax(one_hot, -1), arr))
    return one_hot

下面是一个将一维向量转换为二维单热数组的函数。

#!/usr/bin/env python
import numpy as np

def convertToOneHot(vector, num_classes=None):
    """
    Converts an input 1-D vector of integers into an output
    2-D array of one-hot vectors, where an i'th input value
    of j will set a '1' in the i'th row, j'th column of the
    output array.

    Example:
        v = np.array((1, 0, 4))
        one_hot_v = convertToOneHot(v)
        print one_hot_v

        [[0 1 0 0 0]
         [1 0 0 0 0]
         [0 0 0 0 1]]
    """

    assert isinstance(vector, np.ndarray)
    assert len(vector) > 0

    if num_classes is None:
        num_classes = np.max(vector)+1
    else:
        assert num_classes > 0
        assert num_classes >= np.max(vector)

    result = np.zeros(shape=(len(vector), num_classes))
    result[np.arange(len(vector)), vector] = 1
    return result.astype(int)

下面是一些用法示例:

>>> a = np.array([1, 0, 3])

>>> convertToOneHot(a)
array([[0, 1, 0, 0],
       [1, 0, 0, 0],
       [0, 0, 0, 1]])

>>> convertToOneHot(a, num_classes=10)
array([[0, 1, 0, 0, 0, 0, 0, 0, 0, 0],
       [1, 0, 0, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 0, 1, 0, 0, 0, 0, 0, 0]])

如果你正在使用keras，有一个内置的实用程序:

from keras.utils.np_utils import to_categorical   

categorical_labels = to_categorical(int_labels, num_classes=3)

它与@YXD的答案几乎相同(请参阅源代码)。