如何统计 NumPy 数组中唯一值的出现频率？
让我们看看如何计算 NumPy 数组中唯一值的频率。Python 的 numpy 库提供了 numpy.unique() 函数来查找 numpy 数组中的唯一元素及其对应的频率。
语法: numpy.unique(arr，return_counts=False)
返回:数组中已排序的唯一元素及其对应的频率计数 NumPy 数组。
现在，让我们看看例子:
例 1:
Python 3
# import library
import numpy as np
ini_array = np.array([10, 20, 5,
10, 8, 20,
8, 9])
# Get a tuple of unique values
# and their frequency in
# numpy array
unique, frequency = np.unique(ini_array,
return_counts = True)
# print unique values array
print("Unique Values:",
unique)
# print frequency array
print("Frequency Values:",
frequency)
输出:
Unique Values: [ 5 8 9 10 20]
Frequency Values: [1 2 1 2 2]
例 2:
Python 3
# import library
import numpy as np
# create a 1d-array
ini_array = np.array([10, 20, 5,
10, 8, 20,
8, 9])
# Get a tuple of unique values
# amnd their frequency
# in numpy array
unique, frequency = np.unique(ini_array,
return_counts = True)
# convert both into one numpy array
count = np.asarray((unique, frequency ))
print("The values and their frequency are:\n",
count)
输出:
The values and their frequency are:
[[ 5 8 9 10 20]
[ 1 2 1 2 2]]
例 3:
Python 3
# import library
import numpy as np
# create a 1d-array
ini_array = np.array([10, 20, 5,
10, 8, 20,
8, 9])
# Get a tuple of unique values
# and their frequency in
# numpy array
unique, frequency = np.unique(ini_array,
return_counts = True)
# convert both into one numpy array
# and then transpose it
count = np.asarray((unique,frequency )).T
print("The values and their frequency are in transpose form:\n",
count)
输出:
The values and their frequency are in transpose form:
[[ 5 1]
[ 8 2]
[ 9 1]
[10 2]
[20 2]]
版权属于：月萌API www.moonapi.com，转载请注明出处
