Machine Learning Tutorials, Courses and Certifications
Related Articles
Numpy Array Manipulation¶
In [2]:
import numpy as np
In [3]:
arr = np.random.randint(1,10,[3,3])
arr
Out[3]:
Transpose an array¶
In [4]:
print(arr.T)
or¶
In [5]:
print(np.transpose(arr))
or¶
In [6]:
print(arr.transpose())
numpy.swapaxes() function interchange two axes of an array.¶
- Syntax : numpy.swapaxes(arr, axis1, axis2)
In [7]:
arr1 = np.array([[2, 4, 6]])
arr1
Out[7]:
In [8]:
np.swapaxes(arr1, 0, 1)
Out[8]:
3D Array¶
In [9]:
arr1 = np.arange(16).reshape((2,2,4))
arr1
Out[9]:
In [10]:
np.swapaxes(arr1, 0, 1)
Out[10]:
Change the shape of an array¶
In [11]:
# change the shape of an array and return a copy
arr = np.arange(12)
arr
Out[11]:
In [12]:
arr.reshape((2,6))
Out[12]:
In [13]:
# change the shape of an array in place
arr.resize((2,6))
arr
Out[13]:
flatten an array¶
In [14]:
# return a copy
arr.flatten()
Out[14]:
In [15]:
# return a view
# change any element in the view will change the initial array
arr.ravel()
Out[15]:
append elements to an array¶
In [16]:
arr = np.array([1,2,3])
In [17]:
# append a scalar and return a copy
arr1 = np.append(arr, 4)
print(arr1)
In [18]:
# append an array and return a copy
arr2 = np.append(arr, [4,5,6])
print(arr2)
insert elements into an array¶
In [19]:
# np.insert(array, position, element)
# insert a scalar at a certain position
arr3 = np.insert(arr, 0, 100)
print(arr3)
In [20]:
# insert multiple values at a certain position
arr3 = np.insert(arr, 0, [1,2,3])
print(arr3)
delete elements from an array¶
In [21]:
# remove the element at position 0
arr4 = np.delete(arr, 0)
print(arr4)
In [22]:
# remove the element at multiple positions
arr4 = np.delete(arr, [0,2])
print(arr4)
copy an array¶
In [23]:
arr = np.array([1,2,3])
In [24]:
# the following methods are all deep copy
arr1 = np.copy(arr)
# or
arr1 = arr.copy()
# or
arr1 = np.array(arr, copy=True)
