How to handle Data, and Images(3) - Numpy Usage

Lesson Notes in .ipynb file

How to handle Data, and Images(3) - Numpy Usage

Topics

• Python’s Numpy is capable of saving and loading data
• Saving Multiple datas
• Sorting in Numpy
• More functions that are commonly used with Numpy
• Summary

Python’s Numpy is capable of saving and loading data

Numpy is also capable of saving the any form of data as a seperate file and loading the data as well.

• numpy.save(file, arr, allow_pickle=True, fix_imports=(no value)):
  • saves an array to a binary file in Numpy .npy format
• numpy.load(file, mmap_mode=None, allow_pickle=False, fix_imports=True, encoding=’ASCII’, *, max_header_size=10000):
  • loads arrays or pickled objects from .npy, .npz or pickled files

import numpy as np

# create an array with 0-9
arr = np.arange(0,10)
# save the corresponding array to saved.npy
np.save('saved.npy', arr)

# load the corresponding data to res
res = np.load('saved.npy')
print(res)

Output:

[0 1 2 3 4 5 6 7 8 9]

Saving Multiple datas

Numpy is capable of saving multiple data as well

• numpy.savez(file, *args, allow_pickle=True, **kwds):
  • saves several arrays into a single file in uncompressed .npz format

arr1 = np.arange(0, 10)
arr2 = np.arange(10, 20)
np.savez('saved.npz', arr1=arr1, arr2=arr2)

data = np.load('saved.npz')
res1 = data['arr1']
res2 = data['arr2']
print("res1\n", res1)
print("res2\n", res2)

Output:

res1
 [0 1 2 3 4 5 6 7 8 9]
res2
 [10 11 12 13 14 15 16 17 18 19]

Sorting in Numpy

• numpy.sort(a, axis=-1, kind=None, order=None, *, stable=None):
  • returns a sorted copy of an array

# sorting smallest to biggest
arr = np.array([5, 9, 10, 3, 1])
arr.sort()
print("arr\n", arr)

# sorting biggest to smallest
arr = np.array([5, 9, 10, 3, 1])
arr.sort()
print("arr\n", arr[::-1])

# sorting 2d array by column
arr = np.array([[5, 9, 10, 3, 1], [8, 3, 4, 2, 5]])
arr.sort(axis=0)
print("arr\n", arr)

# sorting 2d array by row
arr = np.array([[5, 9, 10, 3, 1], [8, 3, 4, 2, 5]])
arr.sort(axis=1)
print("arr\n", arr)

Output:

arr
 [ 1  3  5  9 10]
arr
 [10  9  5  3  1]
arr
 [[ 5  3  4  2  1]
 [ 8  9 10  3  5]]
arr
 [[ 1  3  5  9 10]
 [ 2  3  4  5  8]]

More functions that are commonly used with Numpy

• numpy.linspace(start, stop, num=50, endpoint=True, retstep=False, dtype=None, axis=0, *, device=None):
  • returns evenly spaced numbers over a specified interval
• numpy.random.seed(seed=None):
  • reseed the singleton RandomState instance
• numpy.copy(a, order=’K’, subok=False):
  • returns an array copy of the given object
• numpy.unique(ar, return_index=False, return_inverse=False, return_counts=False, axis=None, *, equal_nan=True):
  • finds the unique elements of an array then returns the sorted unique elements of an array.

# create an array for 5 elements from 0 to 10 evenly spaced
arr = np.linspace(0, 10, 5)
print("arr\n", arr)

# random with seed allows us to reenact
np.random.seed(7)
print("random seed (7)\n", np.random.randint(0, 10, (2,3)))

# copying array, it doesn't affect by the changes made to the original array
arr1 = np.arange(0, 10)
arr2 = arr1.copy()
print("arr2\n", arr2)

# removes repeated elements
arr = np.array([1, 1, 2, 3, 3, 3, 1])
print("np.unique\n", np.unique(arr))

Output:

arr
 [ 0.   2.5  5.   7.5 10. ]
random seed (7)
 [[4 9 6]
 [3 3 7]]
arr2
 [0 1 2 3 4 5 6 7 8 9]
np.unique
 [1 2 3]

Summary

• numpy.save: saves an array to a binary file in Numpy .npy format

• numpy.load: loads arrays or pickled objects from .npy, .npz or pickled files

• numpy.savez: saves several arrays into a single file in uncompressed .npz format

• numpy.sort: returns a sorted copy of an array

• numpy.linspace: returns evenly spaced numbers over a specified interval

• numpy.random.seed: reseed the singleton RandomState instance

• numpy.copy: returns an array copy of the given object

• numpy.unique: finds the unique elements of an array then returns the sorted unique elements of an array.