[Numpy] 数组属性

 美图欣赏2022/07/26

---

导入numpy数据包

# 引入Numpy库
import numpy as np

---

1.数组维度数查询

ndarray.ndmin查询数组的维度 

import numpy as np
# 数组维度
## 维度为1
arr1 = np.array([1,2,3])
arr1.ndim # 1
## 维度为2
arr2 = np.array([[1,2,3],[4,5,6]])
arr2.ndim # 2
## 维度为3
arr3 = np.array([
        [[1,2,3],[4,5,6]],
        [[7,8,9],[10,11,12]]
       ])
arr3.ndim # 3

2.数组形状查询  

ndarray.shape查询数组的形状(几行几列)，返回值是一个元组，里面有几个元素代表是几维数组

import numpy as np
arr1 = np.array([1,2,3])
arr1.shape # (3,)

arr2 = np.array([[1,2,3],[4,5,6]])
arr2.shape # (2,3)

arr3 = np.array([
        [[1,2,3],[4,5,6]],
        [[7,8,9],[10,11,12]]
       ])
arr3.shape # (2,2,3)

ndarray.shape也可以改变数组形状

import numpy as np
arr4 = np.array([[1,2,3],[4,5,6]]) 
arr4.shape =  (3,2) 

arr4

arr4(处理后) 

3.修改数组形状 

.reshape函数可以改变原数组的形状，创建一个新数组，改变新数组的元素，原数组对应元素的值也会发生改变 

# NumPy提供了.reshape函数来调整数组大小形状
import numpy as np

data = np.array([[1,2,3],[4,5,6]]) # array([[1, 2, 3],
                                   #        [4, 5, 6]])
data.shape # (2,3)

arr = data.reshape(6,) # array([1, 2, 3, 4, 5, 6])
arr.shape # (6,)

arr[0] = 437
arr # array([437, 2, 3, 4, 5, 6])

data # array([[437,   2,   3],
     #        [  4,   5,   6]])

.flatten函数可实现扁平化(多维数组转化为一维数组)  

import numpy as np

arr = np.array([
        [[1,2,3],[4,5,6]],
        [[7,8,9],[10,11,12]]
      ])
arr.ndim # 3

deal_arr = arr.flatten() # array([1,2,3,4,5,6,7,8,9,10,11,12])
deal_arr.ndim # 1

4.数组元素个数与所占内存查询 

ndarray.size查询数组元素个数

ndarray.itemsize查询数组中每个元素所占内存的大小(以字节为单位)

import numpy as np

arr = np.array([
        [[1,2,3],[4,5,6]],
        [[7,8,9],[10,11,12]]
      ])
# 数组的元素个数
arr.size # 12
# 各元素所占内存
arr.itemsize # 4
# 各元素的数据类型
arr.dtype # dtype('int32')
# 数组所占内存
arr.itemsize * arr.size # 48

# 数组的dtype为int8(一个字节)
data1 = np.array([1,2,3,4,5], dtype = np.int8)
data1.itemsize # 1

# 数组的dtype现在为float64(八个字节)
data2 = np.array([1,2,3,4,5], dtype = np.float64)
data2.itemsize # 8

5.元素数据类型查询

ndarray.dtype用于返回ndarray对象的元素类型 

import numpy as np
arr1 = np.array([[1,2,3],[4,5,6]])  
arr1.dtype # dtype('int32')
arr2 = np.array([1.2, 2.3, 3.4])
arr2.dtype # dtype('float64')