删除 NumPy 数组的行和列
原文:https://www . geesforgeks . org/delete-row-and-columns-of-numpy-ndarray/
在本文中,我们将讨论如何删除 n 维数组中指定的行和列。我们将使用 numpy.delete() 方法删除行和列。
语法: numpy.delete(array_name,obj,axis=None)
让我们借助一些例子来讨论一下:
例 1:
用 NumPy 创建一个二维数组(3 行 4 列)并删除指定的行。
蟒蛇 3
# importing numpy module
import numpy as np
# create an array with integers
# with 3 rows and 4 columns
a = np.array([[1, 2, 3, 4],
[5, 6, 7, 8],
[9, 10, 11, 12]])
print(a)
# delete 0 th row
data = np.delete(a, 0, 0)
print("data after 0 th row deleted :", data)
# delete 1 st row
data = np.delete(a, 1, 0)
print("data after 1 st row deleted :", data)
# delete 2 nd row
data = np.delete(a, 2, 0)
print("data after 2 nd row deleted :", data)
输出:
例 2:
用 NumPy 创建一个二维数组(6 行 2 列)并删除指定的列。
蟒蛇 3
# importing numpy module
import numpy as np
# create an array with integers with
# 6 rows and 2 columns
a = np.array([[1, 2], [5, 6], [9, 10, ],
[78, 90], [4, 89], [56, 43]])
print(a)
# delete 0 th column
data = np.delete(a, 0, 1)
print("data after 0 th column deleted :", data)
# delete 1 st column
data = np.delete(a, 1, 1)
print("data after 1 st column deleted :", data)