I. NumPy Array Creation (数组创建)#

1
import numpy as np
2

3
# 1D Array (一维数组)
4
a = np.array([1, 2, 3, 4, 5])
5
print(a)          # [1 2 3 4 5]
6
print(a.dtype)    # int64
7

8
# 2D Array (二维数组)
9
b = np.array([[1, 2, 3], [4, 5, 6]])
10
print(b)
11
# [[1 2 3]
12
#  [4 5 6]]
13
print(b.shape)    # (2, 3)
14

15
# Specify dtype (指定数据类型)
16
c = np.array([1, 2, 3], dtype=np.float32)
17
print(c)          # [1. 2. 3.]

1
import numpy as np
2

3
# All zeros (全零数组)
4
a = np.zeros((2, 3))
5
print(a)
6
# [[0. 0. 0.]
7
#  [0. 0. 0.]]
8

9
# All ones (全一数组)
10
b = np.ones((3, 2), dtype=int)
11
print(b)
12
# [[1 1]
13
#  [1 1]
14
#  [1 1]]
15

16
# Fill with a specific value (填充指定值)
17
c = np.full((2, 2), 7)
18
print(c)
19
# [[7 7]
20
#  [7 7]]
21

22
# Like versions: same shape as another array (与另一个数组形状相同)
23
d = np.zeros_like(b)
24
print(d)
25
# [[0 0]
26
#  [0 0]]

1
import numpy as np
2

3
# arange: similar to range() but returns ndarray
4
# arange(start, stop, step)
5
a = np.arange(0, 10, 2)
6
print(a)    # [0 2 4 6 8]
7

8
# Float step (浮点步长)
9
b = np.arange(0, 1, 0.3)
10
print(b)    # [0.  0.3 0.6 0.9]
11

12
# linspace: evenly spaced over an interval (等分区间)
13
# linspace(start, stop, num)
14
c = np.linspace(0, 1, 5)
15
print(c)    # [0.   0.25 0.5  0.75 1.  ]
16

17
# logspace: logarithmically spaced (对数等间隔)
18
d = np.logspace(0, 3, 4)   # 10^0 to 10^3
19
print(d)    # [   1.   10.  100. 1000.]

1
import numpy as np
2

3
# Identity Matrix (单位矩阵)
4
a = np.eye(3)
5
print(a)
6
# [[1. 0. 0.]
7
#  [0. 1. 0.]
8
#  [0. 0. 1.]]
9

10
# Offset diagonal (偏移对角线)
11
b = np.eye(3, k=1)
12
print(b)
13
# [[0. 1. 0.]
14
#  [0. 0. 1.]
15
#  [0. 0. 0.]]

1
import numpy as np
2

3
# empty: uninitialized (未初始化, values are random)
4
a = np.empty((2, 2))
5
print(a)    # random values, fast allocation
6

7
# diag: create diagonal matrix or extract diagonal (对角矩阵)
8
b = np.diag([1, 2, 3])
9
print(b)
10
# [[1 0 0]
11
#  [0 2 0]
12
#  [0 0 3]]
13

14
# Extract diagonal from a matrix (提取对角线)
15
c = np.array([[1, 2], [3, 4]])
16
print(np.diag(c))    # [1 4]