AI VIETNAM

Free AI Course 2020

Numpy
Introduction and applications to data
processing

Dinh Quang Vinh

Nguyen Hong Phuc
Year 2020
 AI VIETNAM
Free AI Course 2020

Outline
 Introduction to Numpy
 Numpy Array Indexing
 Numpy Array Operations
 Broadcasting
 Data Processing
 AI VIETNAM
Free AI Course
Introduction
 Numpy is a Python library

Year 2020 1
 AI VIETNAM
Free AI Course
Introduction
 Numpy is a Python library
 For scientific computations

Year 2020 2
 AI VIETNAM
Free AI Course
Introduction
 Numpy array Tensor in
Tensorflow and Pytorch

Year 2020 3
 AI VIETNAM
Free AI Course
Introduction
 Numpy arrays are multi-dimensional arrays

Axis 1

Axis 0 1 2 1 2 3
1 2 3
Axis 0
3 4 Axis 0 4 5 6
5 6 7 8 9
Axis 2

Axis 1

1D array 2D array 3D array

Year 2020 4
 AI VIETNAM
Free AI Course
Introduction
 Create Numpy array
 From List

arr_np = np.array(python_list)

data 0
data
1
0 1
1 2
data 1
2 3
2

Year 2020 5
 AI VIETNAM
Free AI Course
Introduction
 Common attributes
 dtype: data type
 shape: return a tuple of #elements in each dimension
shape=(3,3,2)
 ndim: return #dimensions Axis 1 1 2 3
Axis 0 ndim=2
1 2 4 5 6
Axis 0
1 2 3 shape=(3,) shape=(3,2)
ndim=1 3 4 ndim=2 7 8 9
Axis 2
Axis 0 dtype example
5 6 Axis 1

Year 2020 6
 AI VIETNAM
Free AI Course
Introduction
 Update an element

index 0 1 2
data = 1 2 3

data[0] = 8

data = 8 2 3

Year 2020 7
 AI VIETNAM
Free AI Course
Introduction
 Create Numpy array

zeros() function ones() function full() function

Year 2020 8
 AI VIETNAM
Free AI Course
Introduction
 Create Numpy array
arange() function eye() function random() function
0 1 2 3 4

arr1 = 0 1 2 3 4
0 1 2

arr2 = 0 2 4

Year 2020 9
 AI VIETNAM
Free AI Course
Introduction
 Some important functions
flatten() function
where() function
1 2
arr =
arr = 0 1 2 3 4 3 4
arr<3 = T T T F F flatten()

out = 1 2 3 4
out = 0 1 2 6 8

Year 2020 10
 AI VIETNAM
Free AI Course
Introduction
 Some important functions

reshape() function

data data_rs
1 2 3 1 2
4 5 6 3 4
5 6

Year 2020 11
4
 AI VIETNAM
Free AI Course 2020

Outline
 Introduction to Numpy
 Numpy Array Indexing
 Numpy Array Operations
 Broadcasting
 Data Processing
 AI VIETNAM
Free AI Course
Array Indexing
 Slicing
arr[for_axis_0, for_axis_1, …]
‘:’: get all the elements
‘a:b’: get the elements from a𝑡ℎ to b𝑡ℎ

data data 0, 1 data 1: 3 data 0: 1 , 0 data : , :

0 1 0 1 0 1 0 1 0 1
0 1 2 0 1 2 0 1 2 0 1 2 0 1 2
1 3 4 1 3 4 1 3 4 1 3 4 1 3 4
2 5 6 2 5 6 2 5 6 2 5 6 2 5 6

Year 2020 12
4
 AI VIETNAM
Free AI Course
Array Indexing
 Slicing

Year 2020 13
4
 AI VIETNAM
Free AI Course
Array Indexing
 Slicing
 Mutable

0 1 2
1
0
1 2 3 a_arr [0,1]
a_arr = 0 2
1
5 6 7
b_arr = a_arr[:, 1:3]
0 1 0 1

0 2 3 b_arr [0,1]= 99 0 99 3
b_arr =
1 7 1 6 7
6

result

0 1 2
0
1 99 3
a_arr =
1
5 6 7

Year 2020 14
4
 AI VIETNAM
Free AI Course
Array Indexing
 Get a row

0 1 2

0 1 2 3
arr = 1 5 6 7
2 9 10 11
0 1 2

row_m1 = 5 6 7 shape(3, )
0 1 2

row_m2 = 0 5 6 7 shape(1, 3)

Year 2020 15
4
 AI VIETNAM
Free AI Course
Array Indexing
 Get a column

0 1 2
0
1 2 3
arr = 1
5 6 7
2 9 10 11
0 1 2

col_m1 = 2 6 10 (3, )

0 2
col_m2 = 1 6 (3, 1)
2 10

Year 2020 16
4
 AI VIETNAM
Free AI Course
Array Indexing
 Using Lists as indices

0 1

0 1 2
arr = 1 3 4
2 5 6

arr[[0, 1, 2], [0, 1, 0]] = 1 4 5

arr[[0, 0], [1, 1]] = 2 2

Year 2020 17
4
 AI VIETNAM
Free AI Course
Array Indexing
 Boolean indices

0 1 0 1

0 1 2 0 F F
arr = 1 3 4 arr > 2 = 1 T T
2 5 6 2 T T

0 1 2 3

arr[arr>2] = 3 4 5 6

Year 2020 18
4
 AI VIETNAM
Free AI Course 2020

Outline
 Introduction to Numpy
 Numpy Array Indexing
 Numpy Array Operations
 Broadcasting
 Data Processing
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Addition

data x data y result

1 5 6

2 6 8
+ =
3 7 10

4 8 12

Year 2020 19
4
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Subtraction

data x data y result

1 5 4

2 6 4
- =
3 7 4

4 8 4

Year 2020 20
4
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Multiplication

data x data y result

1 5 5

2 6 12
* =
3 7 21

4 8 32

Year 2020 21
4
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Division

data x data y result

1 5 0.2

2 6 0.33
/ =
3 7 0.42

4 8 0.5

Year 2020 22
4
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Square root

data result

1 1.0

2 1.4
sqrt(data) =
3 1.7

4 2.0

Year 2020 23
4
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Inner product

v w
result
1 2
= 8
2 3

Year 2020 24
4
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Vector-matrix multiplication

X v result

1 2 1 5
=
3 4 2 11

v X result

1 2 1 2 7
=
3 4 10

Year 2020 25
4
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Matrix-matrix multiplication

X Y result

1 2 2 3 6 5
=
3 4 2 1 14 13

Y X result

2 3 1 2 11 16
=
2 1 3 4 5 8

Year 2020 26
4
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Transpose

T
1 2 1 3

3 4 2 4

Year 2020 27
4
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Summation

data
sum(data)
1 2
Axis 0 10
3 4

Axis 1

sum(data, axis=0) sum(data, axis=1)

4 6 3 7

Year 2020 28
4
 AI VIETNAM
Free AI Course
Numpy Array Operations
 Max and min
data data
1 1

2 .max( ) = 3 2 .min( ) = 1

3 3

1 2 1 2
3 4 .max(axis=0) = 5 6 3 4 .min(axis=0) = 1 2
5 6 5 6

1 2 1 2
3 4 .max(axis=1) = 2 4 6 3 4 .min(axis=1) = 1 3 5
5 6 5 6
Year 2020 29
4
 AI VIETNAM
Free AI Course 2020

Outline
 Introduction to Numpy
 Numpy Array Indexing
 Numpy Array Operations
 Broadcasting
 Data Processing
 AI VIETNAM
Free AI Course
Broadcasting
 Vector and a scalar

data result
0 1 2 0 1 2

1 2 3
* 2 = 2 4 6

data result
0 1 2 0 1 2

1 2 3 - 2 = -1 0 1

Year 2020 30
4
 AI VIETNAM
Free AI Course
Broadcasting
 Matrix and vector
X
0 1 2
0
1 2 3
v
1
4 5 6
2 7 8 9
+ 1 0 1

3 10 11 12

Y
0 1 2 0 1 2
0
1 2 3 1 0 1 0
2 2 4
1
4 5 6 1 0 1 1
5 5 7
2 7 8 9
+ 1 0 1
=
2 8 8 10
3 10 11 12 1 0 1 3 11 11 13
Year 2020 31
4
 AI VIETNAM
Free AI Course 2020

Outline
 Introduction to Numpy
 Numpy Array Indexing
 Numpy Array Operations
 Broadcasting
 Data Processing
 AI VIETNAM
Free AI Course
Data Processing
 Text data
 IRIS data

Year 2020 32
4
 AI VIETNAM
Free AI Course
Data Processing
 Text data
 IRIS data

col 0 col 1 col 2 col 3 col 4

Demo

Year 2020 33
4
 AI VIETNAM
Free AI Course
Image Data
 Grayscale images

(Height, Width)
Pixel p = scalar Resolution: #pixels
Resolution = HeightxWidth
0 ≤ p ≤ 255
Year 2020 http://jalammar.github.io/visual-numpy/ 34
4
 AI VIETNAM
Free AI Course
Image Classification: Image Data
 Color images

(Height, Width, channel)

𝑟
RGB color image Pixel p= 𝑔 Resolution: #pixels
𝑏 Resolution = HeightxWidth

Year 2020 0 ≤ r,g,b ≤ 255 http://jalammar.github.io/visual-numpy/ 35

4
Image Data T-shirt

Trouser
Fashion-MNIST dataset
Pullover

Dress
Grayscale images

Resolution=28x28 Coat

Training set: 60000 samples Sandal

Testing set: 10000 samples

Shirt
28
Sneaker
28

Bag

Ankle
784 Boot
 AI VIETNAM
Free AI Course
OpenCV
 Introduction

Year 2020 37
4
 AI VIETNAM
Free AI Course 2020

Summary
Indexing
Broadcasting data data 0, 1 data 1: 3 data 0: 1 , 0 data : , :
0 1 0 1 0 1 0 1 0 1
0 1 2 0 1 2 0 1 2 0 1 2 0 1 2
1 3 4 1 3 4 1 3 4 1 3 4 1 3 4
2 5 6 2 5 6 2 5 6 2 5 6 2 5 6

Multi-dimension Array

1 2 3
Axis 0 4 5 6
7 8 9
Axis 2

Axis 1
38