简单来说,to_categorical就是将类别向量转换为二进制(只有0和1)的矩阵类型表示。其表现为将原有的类别向量转换为独热编码的形式。先上代码看一下效果:
from keras.utils.np_utils import *
#类别向量定义
b = [0,1,2,3,4,5,6,7,8]
#调用to_categorical将b按照9个类别来进行转换
b = to_categorical(b, 9)
print(b)
执行结果如下:
[[1. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 1. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 1. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 1. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 1. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 1. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 1. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 1. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 1.]]
to_categorical最为keras中提供的一个工具方法,从以上代码运行可以看出,将原来类别向量中的每个值都转换为矩阵里的一个行向量,从左到右依次是0,1,2,...8个类别。2表示为[0. 0. 1. 0. 0. 0. 0. 0. 0.],只有第3个为1,作为有效位,其余全部为0。
独热编码又称为一位有效位编码,上边代码例子中其实就是将类别向量转换为独热编码的类别矩阵。也就是如下转换:
0 1 2 3 4 5 6 7 8
0=> [1. 0. 0. 0. 0. 0. 0. 0. 0.]
1=> [0. 1. 0. 0. 0. 0. 0. 0. 0.]
2=> [0. 0. 1. 0. 0. 0. 0. 0. 0.]
3=> [0. 0. 0. 1. 0. 0. 0. 0. 0.]
4=> [0. 0. 0. 0. 1. 0. 0. 0. 0.]
5=> [0. 0. 0. 0. 0. 1. 0. 0. 0.]
6=> [0. 0. 0. 0. 0. 0. 1. 0. 0.]
7=> [0. 0. 0. 0. 0. 0. 0. 1. 0.]
8=> [0. 0. 0. 0. 0. 0. 0. 0. 1.]
那么一道思考题来了,让你自己编码实现类别向量向独热编码的转换,该怎样实现呢?
以下是我自己粗浅写的一个小例子,仅供参考:
def convert_to_one_hot(labels, num_classes):
#计算向量有多少行
num_labels = len(labels)
#生成值全为0的独热编码的矩阵
labels_one_hot = np.zeros((num_labels, num_classes))
#计算向量中每个类别值在最终生成的矩阵“压扁”后的向量里的位置
index_offset = np.arange(num_labels) * num_classes
#遍历矩阵,为每个类别的位置填充1
labels_one_hot.flat[index_offset + labels] = 1
return labels_one_hot
#进行测试
b = [2, 4, 6, 8, 6, 2, 3, 7]
print(convert_to_one_hot(b,9))
测试结果:
[[0. 0. 1. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 1. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 1. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 1.]
[0. 0. 0. 0. 0. 0. 1. 0. 0.]
[0. 0. 1. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 1. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 1. 0.]]
to_categorical在keras的utils/np_utils.py中,源码如下:
def to_categorical(y, num_classes=None, dtype='float32'):
"""Converts a class vector (integers) to binary class matrix.
E.g. for use with categorical_crossentropy.
# Arguments
y: class vector to be converted into a matrix
(integers from 0 to num_classes).
num_classes: total number of classes.
dtype: The data type expected by the input, as a string
(`float32`, `float64`, `int32`...)
# Returns
A binary matrix representation of the input. The classes axis
is placed last.
# Example
```python
# Consider an array of 5 labels out of a set of 3 classes {0, 1, 2}:
> labels
array([0, 2, 1, 2, 0])
# `to_categorical` converts this into a matrix with as many
# columns as there are classes. The number of rows
# stays the same.
> to_categorical(labels)
array([[ 1., 0., 0.],
[ 0., 0., 1.],
[ 0., 1., 0.],
[ 0., 0., 1.],
[ 1., 0., 0.]], dtype=float32)
```
"""
#将输入y向量转换为数组
y = np.array(y, dtype='int')
#获取数组的行列大小
input_shape = y.shape
if input_shape and input_shape[-1] == 1 and len(input_shape) > 1:
input_shape = tuple(input_shape[:-1])
#y变为1维数组
y = y.ravel()
#如果用户没有输入分类个数,则自行计算分类个数
if not num_classes:
num_classes = np.max(y) + 1
n = y.shape[0]
#生成全为0的n行num_classes列的值全为0的矩阵
categorical = np.zeros((n, num_classes), dtype=dtype)
#np.arange(n)得到每个行的位置值,y里边则是每个列的位置值
categorical[np.arange(n), y] = 1
#进行reshape矫正
output_shape = input_shape + (num_classes,)
categorical = np.reshape(categorical, output_shape)
return categorical
看过源码之后,确实觉得自己的代码还需要完善。框架里的一些api,我们可以先自己想着来写,然后和源码进行对比学习,这是一个很好的学习方法。