1 of 61

AIBridge

Lecture 1

2 of 61

AIBridge

Bridge the gap between AI and [your choice]
1^st camp at UC Davis in June 2022, 2^nd in Silicon Valley in March 2023, 3^rd at Cornell in July 2023
Acquire basics: Python, basic ML algorithms, toolbox usage
Enable further learning
Enable easier communications and collaborations

AIFS - NSF/USDA AI Institute for Next Generation Food Systems

3 of 61

AI in Food Systems

Molecular breeding

Help breeders to run more efficient and targeted breeding programs

Agricultural production

Crop yield sensing and forecasting
Water and nitrogen stress sensing, prediction, accusation

Food processing

Tomato processing loss prediction
Sanitation classification

Nutrition

Use food photo and text to predict core ingredients
Dietary recommendation

4 of 61

WHAT IS AI/ML?

4

AI vs. ML

I understand that many of you are here because of the popularity of AI. So what is AI? To answer this seemingly simple question is actually fairly complicated. At its simplest form, artificial intelligence is a field, which combines computer science and robust datasets, to enable problem-solving. Some definitions also include ”mimic human cognitive functions”. Through AI, a computer system uses math and logic to simulate the reasoning that people use to learn from new information and make decisions.

In this bootcamp, although we called it AIBridge. Our focus really is Machine learning. Our focus really is Machine learning. ML is a subset of AI which allows a machine to automatically learn from past data without programming explicitly.

We will use AI/ML interchangeablely in this camp. Do note there is a subtle difference between the two.

Following this line of thought of learning from past data, let’s look at what it means more precisely.

An “intelligent” computer uses AI to think like a human and perform tasks on its own. Machine learning is how a computer system develops its intelligence.

https://azure.microsoft.com/en-us/overview/artificial-intelligence-ai-vs-machine-learning/#introduction

5 of 61

What can AI do

6 of 61

A High-Level View

6

7 of 61

Deep Learning

7

8 of 61

8

9 of 61

Our focus

10 of 61

Terminology alignment

Sample = (features, label)
Features: independent variables, attributes, predictors, input variables, input, covariates, explanatory variables, treatment variables,
Label: dependent variable, outcome, target variable, outcome variable, response variable
Samples: cases, observations, individuals, participants, data points

If you have other names for these, please let me know.

11 of 61

Class Structure

Two blocks a day
A block : Lecture + break + lab

Lab is the best part of this bootcamp

Recap

Overview of key knowledge points
Feedback from you (pace, clarity, etc.)

Learning by doing

Iris dataset
Wine dataset

Go through the process to complete a basic ML project

12 of 61

Schedule

Python: 1.5 days

Condensed with a focus on what we need for ML

ML: 3.5 days

Intuitions, toolbox, ML flow

Friday afternoon: Project presentation

13 of 61

Schedule

14 of 61

Typical Practices in ML/Programming

Find a sample
Read through it
Try it
Modify it
Google it

Basic skills to do these and practice them

15 of 61

Best Practices

Ask questions
Type along during lectures
Ask for help
Make good use of labs
Provide feedback

16 of 61

Learning by Doing

Iris
Wine
Your own on Day 5 PM

17 of 61

Resources

Class notes, links in notes
Python: https://www.w3schools.com/python/
Sklearn user guide: https://scikit-learn.org/stable/user_guide.html
Google
ChatGPT*

18 of 61

INTRODUCTION TO PYTHON

19 of 61

Python

Python is a popular programing language
Guido van Rossum, Dutch programmer, invented in late 1980s
Widely used in industry and academia, especially for ML applications.
R vs Python

Python better at large data amounts and machine learning

When he began implementing Python, Guido van Rossum was also reading the published scripts from “Monty Python's Flying Circus”, a BBC comedy series from the 1970s. Van Rossum thought he needed a name that was short, unique, and slightly mysterious, so he decided to call the language Python.

Reminder to talk about how much python is used:

Web development, data analytics and machine learning, automation, scripting, etc. It is beginner friendly with simple syntax, versatile for many different tasks, open source, lot of modules and libraries.

Python and R are the preferred languages in Data Science, Data Analysis, Machine Learning, etc. Although they are used for similar purposes they differ from each other. R mainly focuses on the statistical part of a project while Python is flexible in its usage and data analysis tasks and has more ML libraries.

R is a powerful tool for visualizing data in the form of graphs. It is difficult to use R in a production environment due to its yet-to-develop production tools, whereas Python is easily integrated with a complex work environment.

Considering performance, Python is a better option as it runs faster in all the environments than R.

Python has an easier learning curve.

�

20 of 61

Lecture Outline

20

Control Flows

Logic

Variables

General Python Syntax

Google Colab

21 of 61

https://colab.research.google.com/
Stores everything on Google Drive
Can be shared with others and across devices
No setup required
Most packages/libraries preinstalled

Colab Pro

Follow along as we work through the Python language

21

Google Colab

22 of 61

22

Write code inside “cells”

Run a cell by clicking this button

Add new cells by clicking “+ Code”

Delete cells by clicking this button

Google Colab

23 of 61

Lecture Outline

23

Control Flows

Logic

Variables

General Python Syntax

Google Colab

24 of 61

Comments allow sections of the code to be more readable

Anything after a “#” is a comment
# I am a comment!

Functions take in inputs and give outputs

print(input)

The print function prints out the input

print("hello world")

24

General Python Syntax

Getting Started

We’re going to first talk about some general python syntax that will be helpful for any piece of code you write: Commenting and Functions

Comments are technically optional but they are useful to make your code readable to both yourself and others. In order to make a comment, just type a # symbol followed by the contents of the comment. You will know when something is commented because it will turn green. I generally use comments to…

Functions will be discussed more in depth in the future, but for now, the most relevant information to remember about functions is how to use a function, or calling a function. First, you put the name of the function followed by open parentheses and then any inputs followed by closed parentheses. I’m going to introduce the print function. This function will take whatever input you give and print it to the screen.

Let’s switch over to the colab screen. (demo hello world with a comment).

25 of 61

Lecture Outline

Google Colab

25

General Python Syntax

Variables

Logic

Control Flows

26 of 61

A variable is a reserved place in memory (think: container) which can store a value

Creating variables: variable_name = value

Can be used anywhere after its assignment, but never before
Can re-assign values as needed
7 types of values: Integer, Floating-point, String, Boolean, List, Tuple, and Dictionary

(More details about each type coming up in next slides)

26

var_a = 25

var_a = 70

print(var_a)

Variables

Overview

27 of 61

Cannot start with a number
Cannot include spaces
Cannot be a keyword: https://www.w3schools.com/python/python_ref_keywords.asp
Should be descriptive
*Good practice: all lowercase with underscores for spacing

Good examples: datapoint_number, petal_width, ...

27

"3rd_variable"

"my variable"

Variables

Names

Let’s go over some rules involving variable names. There are a couple hard rules that must be followed otherwise your program will break. There are also some soft rules that are just good practice.

These first three are hard rules so you must follow them.

The first rule is that variables cannot start with a number. This is to prevent the computer from confusing a variable with a number.

The second rule is that variables cannot have spaces. Use underscores instead.

The third rule is variable names cannot be the same as a keyword that’s already used by Python and is reserved for other purposes.

Additionally, these last two are soft rules that are good practice.

Variable names should be descriptive. You will likely be using the same variables over and over again, so giving them a good name makes it easy to remember what their purpose is.

Next up, variables should follow the snake case convention. This means they should be composed of all lower case letters with underscores for spacing.

28 of 61

70 ⇒ because the value of variable_b is set to be 70 in the second line
40 ⇒ because the value of variable_b is set to be the same as variable_a which is 40
25 ⇒ because the value of variable_b is set to be the same as variable_a which is 25

What does the following code output?

variable_a = 25

varaible_b = 70

variable_a = 40

variable_b = variable_a

print(variable_b)

Variables

Self-test

29 of 61

70 ⇒ because the value of variable_b is set to be 70 in the second line
40 ⇒ because the value of variable_b is set to be the same as variable_a which is 40
25 ⇒ because the value of variable_b is set to be the same as variable_a which is 25

What does the following code output?

variable_a = 25

varaible_b = 70

variable_a = 40

variable_b = variable_a

print(variable_b)

Variables

Self-test

30 of 61

Non-fractional number
Positive or negative
No maximum or minimum practically

first_number = 1

second_number = 5

third_number = -3

30

Variables

Integer

31 of 61

“Float”
Decimal point number
Accurate within 2^-55

petal_length = 3.5

petal_width = 4.0

pi = 3.14159265358

31

3.1415926

Floating (Decimal) Point

Variables

Floating-point

32 of 61

A string of characters
Put in quotations " " or ' '
*Block string (multi-line string): three quotation marks
*Special character (new line): "\n"

first_string = "s"

second_string = "string 2"

second_string = "another string"

32

Not this

Variables

String

33 of 61

True or False (capitalize)

first_boolean = True

second_boolean = False

33

Variables

Boolean

34 of 61

A list of values

my_list = [value_1, value_2, ...]
example_list = [5, 20, 11, 3, 10]
Can include multiple different data types

multi_type_list = ["hello world", True, 5]

34

[a, b, c, d, e]

0 1 2 3 4

For a specific value in the list: my_list[index]

The index of the 1st item is 0,
a_value = my_second_list[2] # gets the THIRD value in the list
There is also negative indexing (index of -1 gets last element, -2 gets second from last, etc.)

Variables

List

35 of 61

What does the following code output?

22 ⇒ because value is set to the second item in the list
23 ⇒ because value is set to the third item in the list

my_list = [21, 22, 23, 24, 25]

value = my_list[2]

print(value)

Variables

Self-test

36 of 61

What does the following code output?

22 ⇒ because value is set to the second item in the list
23 ⇒ because value is set to the third item in the list

my_list = [21, 22, 23, 24, 25]

value = my_list[2]

print(value)

Variables

Self-test

37 of 61

Works the same as a list, but can’t be changed
Can contain multiple different data types

my_first_tuple = (object_1, object_2, ...)

my_second_tuple = (22, "hello!", True, 3.1415)

a_value = my_second_tuple[2] # gets the THIRD value in the tuple

37

Variables

* Tuple

38 of 61

A list of values with custom keys that are indices, like a list but indices are keys and not positions

my_dictionary={'apple':'fruit', 'banana':'fruit', 'cabbage':'vegetable', 'dragonfruit':'fruit','eggplant':'vegetable'}

print(my_dictionary['cabbage'])

38

Variables

* Dictionary

39 of 61

Types: int, float, str, bool, list, tuple
Convert types of variables to other types

my_float = float(my_string) #gives string in float form if possible

Compatible types:

int → float
float → int (always rounds down)
str → int
str → float
*[most types] → string
*list → tuple
*boolean → int/float (0 → False, anything else → True)
*str → list/tuple (only converts str to list/tuple of single characters)

39

Variables

Type Conversion

40 of 61

+	-	*	**	/	//	%
Addition	Subtraction	Multiplication	Exponentiation	Division (turns int to float)	Floor Division (rounds down the quotient)	Modulus (returns the remainder)
x + y 1 + 2 == 3	x - y 2 – 1 == 1	x * y 2 * 3 == 6	x y 2 3 == 8	x / y 8 / 2 == 4.0	x // y 9 // 4 == 2	x % y 10 % 4 == 2

Note: the double equal sign a == b is used to check for equality instead of assigning variables

Variables

Basic Arithmetic Operations

41 of 61

x = 4

x = x + 1

# x becomes 5

Changing a variable’s value:

x = 4

x = x - 2

# x becomes 2

x = 4

x = x * 2

# x becomes 8

Variables

Basic Arithmetic Operations

42 of 61

Lecture Outline

42

Control Flows

Logic

Variables

General Python Syntax

Google Colab

43 of 61

if statement_1:

Code segment 1

elif statement_2: # elif means else if

Code segment 2

else:

Code segment 3

43

Logic

Conditionals

44 of 61

x =

y =

if x == y:

print('x is equal to y')

elif x > y:

print('x is greater than y')

else:

print('x is less than y')

Logic

Conditionals

1

45 of 61

x = 4

y =

if x == y:

print('x is equal to y')

elif x > y:

print('x is greater than y')

else:

print('x is less than y')

Logic

Conditionals

1

46 of 61

x = 4

y =

if x == y:

print('x is equal to y')

elif x > y:

print('x is greater than y')

else:

print('x is less than y')

Logic

Conditionals

10

47 of 61

== != < > <= >=

== True if the two sides are exactly the same (1 == 1 is True)

!= True if the two sides are NOT the same (2 != 1 is True)

47

Logic

Conditionals

48 of 61

and: only runs if both are True

if 1 == 1 and 1 == 2:

code segment…

or: runs if at least one of them are True

if 1 == 1 or 1 == 2:

code segment…

48

x = 4

y = 4

if x < y or x == y:

print("x is less than or equal to y")

Logic

Conditionals

49 of 61

Which of these conditions are successfully passed?

petal_width = 1.8

petal_length = 3.5

if petal_width < 3 or petal_length < 3:

print("condition 1 passed")

if petal_width < 3 and petal_length < 3:

print("condition 2 passed")

if petal_width < 3:

if petal_length < 3:

print("condition 3 passed")

Logic

Self-Test

50 of 61

petal_width = 1.8

petal_length = 3.5

if petal_width < 3 or petal_length < 3:

print("condition 1 passed")

if petal_width < 3 and petal_length < 3:

print("condition 2 passed")

if petal_width < 3:

if petal_length < 3:

print("condition 3 passed")

Which of these conditions are successfully passed?

Logic

Self-Test

51 of 61

Lecture Outline

51

Control Flows

Logic

Variables

General Python Syntax

Google Colab

52 of 61

We have this very large list of 11 words:

How do we access and print out every word?

52

word_list = ["Lorem", "ipsum", "dolor", "sit", "amet", "fusce", "rhoncus", "mi", "viverra", "velit", "mattis"]

Control Flows

Hypothetical Scenario

53 of 61

print(word_list[0])

print(word_list[1])

print(word_list[2])

print(word_list[3])

print(word_list[4])

print(word_list[5])

print(word_list[6])

print(word_list[7])

print(word_list[8])

print(word_list[9])

print(word_list[10])

Horribly inefficient

A lot of tedious manual coding

Completely unscalable (what if there were 70 words)

53

word_list = ["Lorem", "ipsum", "dolor", "sit", "amet", "fusce", "rhoncus", "mi", "viverra", "velit", "mattis"]

Control Flows

Hypothetical Scenario

54 of 61

Only difference between all these lines is the index

54

print(word_list[0])

print(word_list[1])

print(word_list[2])

print(word_list[3])

print(word_list[4])

print(word_list[5])

print(word_list[6])

print(word_list[7])

print(word_list[8])

print(word_list[9])

print(word_list[10])

Control Flows

Hypothetical Scenario

55 of 61

How to use: for iterator in iterable:

String, list, range, etc. iterable
Need indentation

55

for number in range(0, 11): #range goes through 0, 1, 2, … 10

#this loop repeats 11 times and number changes to each number

print(word_list[number])

word_list = ["Lorem", "ipsum", "dolor", "sit", "amet", "fusce", "rhoncus", "mi", "viverra", "velit", "mattis"]

Control Flows

For loops

56 of 61

for word in word_list:

#this loop does the exact same thing but with less typing

print(word)

56

for number in range(0, 11): #range goes through 0, 1, 2, … 10

#this loop repeats 11 times and number changes to each number

print(word_list[number])

word_list = ["Lorem", "ipsum", "dolor", "sit", "amet", "fusce", "rhoncus", "mi", "viverra", "velit", "mattis"]

Control Flows

For loops

57 of 61

word_list = ["Lorem", "ipsum", "dolor", "sit", "amet", "fusce", "rhoncus", "mi", "viverra", "velit", "mattis"]

57

Output:

for word in word_list:

#this loop does the exact same thing but with less typing

print(word)

Control Flows

Self-test

58 of 61

a.

for word in big_list:

print(word)

b.

for i in range(9):

print(big_list[i])

c.

for word in big_list:

print(big_list[word])

Which of the following code blocks will print out everything in the list?

big_list = ["Lorem", "Ipsum", "Dolor", "Sit", "Amet", "Consectetur", "Adipiscing", "Elit", "Sed"]

Control Flows

Self-test

59 of 61

Which of the following code blocks will print out everything in the list?

big_list = ["Lorem", "Ipsum", "Dolor", "Sit", "Amet", "Consectetur", "Adipiscing", "Elit", "Sed"]

a.

for word in big_list:

print(word)

b.

for i in range(9):

print(big_list[i])

c.

for word in big_list:

print(big_list[word])

Control Flows

Self-test

60 of 61

How to use: while statement:

The loop repeats as statement is true
Needs indentation

my_number = 0

while my_number < 6:

print(my_number)

my_number = my_number + 1

60

Control Flows

While

61 of 61

a_list = [3, 22, 1, 73, 40, 3, 19]

sum = 0

for i in range(0, 7):

sum = sum + a_list[i]

sum = sum / 2.4

sum = sum * -1

print(a_list[i])

print(sum)

Inside loop because of indentation

(tab)

61

Don’t worry about what this code does.

Control Flows

Indentation