Introduction to Python

Part 1
v0.7

Research Computing Services

Information Services & Technology
RCS Team and Expertise

 Our Team  Consulting Focus:

 Scientific Programmers  Bioinformatics
 Systems Administrators  Data Analysis / Statistics
 Graphics/Visualization Specialists  Molecular modeling
 Account/Project Managers  Geographic Information Systems
 Special Initiatives (Grants)  Scientific / Engineering Simulation
 Maintains and administers the Shared  Visualization
Computing Cluster
 Located in Holyoke, MA
 ~19,000 CPUs running Linux
 CONTACT US: [email protected]
SCC Python Tutorials
Intermediate/Advanced Topics

 Introduction to Python  Introduction to PyTorch

 Numerical Python  Multi-PyTorch
 Introduction to Python Pandas  Python Optimization
 Pretty Python Coding  Python Parallelization
About You

 Working with Python already?

 Have you used any other programming languages?

 Why do you want to learn Python?

Python on the SCC

 There is documentation for using Python on the SCC on the RCS website.
 Use the module system to find Python:
Getting Python for Yourself: Anaconda

 The most popular setup for personal computers

 https://www.anaconda.com/download/

 Anaconda is a packaged set of programs including the Python language,

a huge number of libraries, and several tools.
 These include the Spyder development environment and Jupyter notebooks.

 Anaconda can be used on the SCC, with some setup required.

Python 2 vs. 3
 Python 2: released in 2000, Python 3 released in 2008
 Python 2 is in “end-of-life” mode – no new releases or patches after Dec 31, 2019
 No library support (e.g. pandas, numpy, etc.) after that date
 For more: https://www.python.org/doc/sunset-python-2/

 Py3 is not completely compatible with Py2

 For learning Python these differences are almost negligible

 Which one to learn?

 Python 3
 If your research group / advisor / boss / friends all use one version that’s probably the best one
for you to choose.
 If you have a compelling reason to focus on one vs the other
Spyder – a Python development environment

 Pros:
 Faster development
 Easier debugging!
 Helps organize code
 Increased efficiency

 Cons
 Learning curve
 Can add complexity to smaller
problems
The Spyder IDE

Variable and file explorer

file editor

Python console
Tutorial Outline – Part 1

 What is Python?
 Operators
 Variables
 Functions
 Classes
 If / Else
Tutorial Outline – Part 2

 Lists
 Tuples and dictionaries
 Modules
 numpy and matplotlib modules
 Script setup
 Development notes
Tutorial Outline – Part 1

 What is Python?
 Operators
 Variables
 Functions
 Classes
 If / Else
What is Python?

 Python…
 …is a general purpose interpreted programming language.
 …is a language that supports multiple approaches to software design,
principally structured and object-oriented programming.
 …provides automatic memory management and garbage collection
 …is extensible
 …is dynamically typed.

 By the end of the tutorial you will understand all of these terms.
Some History
 “Over six years ago, in December 1989, I was looking for a "hobby"
programming project that would keep me occupied during the week
around Christmas…I chose Python as a working title for the project, being
in a slightly irreverent mood (and a big fan of Monty Python's Flying
Circus).”
–Python creator Guido Van Rossum, from the foreward to Programming Python (1st ed.)

 Goals:
 An easy and intuitive language just as powerful as major competitors
 Open source, so anyone can contribute to its development
 Code that is as understandable as plain English
 Suitability for everyday tasks, allowing for short development times
Compiled Languages (ex. C++ or Fortran)
Interpreted Languages (ex. Python or R)

bytecode Python interpreter:

Source code files
Python interpreter follows bytecode
prog.py conversion
math.py instructions

python prog.py

 A lot less work is done to get a program to start running compared with compiled
languages!
 Python programs start running immediately – no waiting for the compiler to finish.
 Bytecodes are an internal representation of the text program that can be efficiently run by
the Python interpreter.
 The interpreter itself is written in C and is a compiled program.
 Comparison

Interpreted (Python, R, Matlab…) Compiled (C++, Julia, Fortran…)

 Faster development  Longer development
 Easier debugging  Edit / compile / test cycle is longer!
 Debugging can stop anywhere, swap in  Harder to debug
new code, more control over state of  Usually requires a special compilation
program
 (almost always) takes more code to
 (almost always) takes less code to get things done
get things done
 Faster
 Slower programs  Compiled code runs directly on CPU
 Sometimes as fast as compiled, rarely  Compiler can optimize more
faster extensively
 Less control over program behavior  More control over program behavior
17
The Python Prompt
 The standard Python prompt looks like this:

 The IPython prompt in Spyder looks like this:

 IPython adds some handy behavior around the standard Python prompt.
Tutorial Outline – Part 1

 What is Python?
 Operators
 Variables
 Functions
 Classes
 If / Else
Operators

 Python supports a wide variety of operators which act like functions, i.e.
they do something and return a value:
 Arithmetic: + - * / % **
 Logical: and or not
 Comparison: > < >= <= != ==
 Assignment: =
 Bitwise: & | ~ ^ >> <<
 Identity: is is not
 Membership: in not in
Try Python as a calculator

 Go to the Python prompt.

 Try out some arithmetic operators:

+ - * / % ** == ( )

 Can you identify what they all do?

Try Python as a calculator

 Go to the Python prompt.

 Try out some arithmetic operators:
+ - * / % ** == ()
Operator Function
+ Addition
- Subtraction
* Multiplication
/ Division (Note: 3 / 4 is 0.75!)
% Remainder (aka modulus)
** Exponentiation
== Equals
More Operators
 Try some comparisons and Boolean operators. True and False are the
keywords indicating those values:
Comments
 # is the Python comment character. On
any line everything after the # character
is ignored by Python.

 There is no multi-line comment

character as in C or C++.

 An editor like Spyder makes it very easy

to comment blocks of code or vice-
versa. Check the Edit menu
Tutorial Outline – Part 1

 What is Python?
 Operators
 Variables
 Functions
 Classes
 If / Else
Variables

 Variables are assigned values using the = operator

 In the Python console, typing the name of a variable
prints its value
 Not true in a script!
 Visualize Assignment

 Variables can be reassigned at any time

 Variable type is not specified
 Types can be changed with a reassignment
Variables cont’d

 Variables refer to a value stored in memory and are created when first
assigned
 Variable names:
 Must begin with a letter (a - z, A - Z) or underscore _
 Other characters can be letters, numbers or _
 Are case sensitive: capitalization counts!
 Can be any reasonable length
 Assignment can be done en masse:
x = y = z = 1
Try these out!
 Multiple assignments can be done on one line:
x, y, z = 1, 2.39, 'cat'
Variable Data Types

 Python determines data types for variables based on the context

 The type is identified when the program runs, using dynamic typing
 Compare with compiled languages like C++ or Fortran, where types are identified by
the programmer and by the compiler before the program is run.

 Run-time typing is very convenient and helps with rapid code development

 But… it requires the programmer to do more code testing for reliability.

Variable Data Types
 Available basic types:
 Numbers: Integers and floating point (64-bit)
 Complex numbers: x = complex(3,1) or x = 3+1j
 Strings, using double or single quotes: "cat" 'dog'
 Boolean: True and False
 Lists, dictionaries, sets, and tuples
 These hold collections of variables
 Specialty types: files, network connections, objects

 Custom types can be defined using Python classes.

Variable modifying operators

 Some additional arithmetic operators that modify variable values:

Operator Effect Equivalent to…
x += y Add the value of y to x x=x+y
x -= y Subtract the value of y x =x-y
from x
x *= y Multiply the value of x x =x*y
by y
x /= y Divide the value of x by x =x/y
y

 The += operator is by far the most commonly used of these.

Check a type

 A built-in function, type(), returns the

type of the data assigned to a variable.
 It’s unusual to need to use this in a program, but it’s
available if you need it.

 Try this out in Python – do some

assignments and reassignments and
see what type() returns.

31
Strings
 Strings are a basic data type in Python.

 Indicated using pairs of single '' or

double "" quotes.
 Multiline strings use a triple set of
quotes (single or double) to start and
end them.

32
Tutorial Outline – Part 1

 What is Python?
 Operators
 Variables
 Functions
 Classes
 If / Else
Functions
 Functions are used to create pieces of code that can be used in a
program or in many programs.

 The use of functions is to logically separate a program into discrete

computational steps.

 Programs that make heavy use of function definitions tend to be easier to:
 develop
 debug
 maintain
 understand
 Function name
Python functions Optional comma-separated
list of arguments (incoming
variables)

Keyword def
A code block

Optional return statement

 The return value can be any Python type

 If the return statement is omitted a special None value is still returned.
 The arguments are optional but the parentheses are required!
 Functions must be defined before they can be called.
Write a Function

 In Spyder’s editor:

 Define a function called mathcalc that takes 3 numbers as arguments and returns
their sum divided by their product.

 Save the file, run it, and then call it from the console:

mathcalc(1,2,3)  returns 1.0

mathcalc(4, -2.5, 3.0)  returns -0.15

 Which code sample is easier to read?

float avg(int a, int b, int c)

float avg(int a, int b, int c){ {
 C: float sum = a + b + c ; or float sum = a + b + c ;
return sum / 3.0 ;} return sum / 3.0 ;
}

 Matlab: function a = avg(x,y,z) function a = avg(x,y,z)

a = x + y + z ; or a = x + y + z ;
a = a / 3.0 ; a = a / 3.0 ;
end end
 Which code sample is easier to read?
C
float avg(int a, int b, int c)
 Most languages use special characters ({ } {
float sum = a + b + c ;
pairs) or keywords (end, endif) to indicate return sum / 3.0 ;
sections of code that belong to: }
 Functions
 Control statements like if
 Loops like for or while Matlab
function a = avg(x,y,z)
a = x + y + z ;
 Python instead uses the indentation that a = a / 3.0 ;
programmers use anyway for readability. end
The Use of Indentation
 Python uses whitespace (spaces or tabs) to define code blocks.
 Code blocks are logical groupings of commands. They are always
preceded by a colon :
def avg(x,y,z): A code block
all_sum = x + y + z
return all_sum / 3.0

def mean(x,y,z):
Another code block return (x + y + z) / 3.0

 This pattern is consistently repeated throughout Python syntax.

 Spaces or tabs can be mixed in a file but not within a code block.
Function Return Values

 A function can return any Python value.

 Function call syntax:

A = some_func() # some_func returns a value
Another_func() # ignore return value or nothing returned
b,c = multiple_vals(x,y,z) # return multiple values

 Open function_calls.py for some examples

Function arguments
 Function arguments can be required or optional.
 Optional arguments are given a default value

def my_func(a,b,c=10,d=-1):
…some code…

 To call a function with optional arguments:

 Optional arguments can be used in the order they’re declared or out of
order if their name is used.

my_func(x,y,z) # a=x, b=y, c=z, d=-1

my_func(x,y) # a=x, b=y, c=10, d=-1
my_func(x,y,d=w,c=z) # a=x, b=y, c=z, d=w
Garbage collection
 Variables defined in a function no longer have any “live” references to
them once the function returns.

 These variables become garbage – memory that can no longer be used.

 Python’s garbage collection operates to return that memory to the

operating system so it can be re-used.

 There is no need to explicitly destroy or release most variables.

 Some complex data types provide .close(), .clean(), etc. type functions. Use these where
available.
When does garbage collection occur?
 It happens when Python thinks it should.
 Part of the process is based on heuristics, i.e. tuned parameters in Python.
 For the great majority of programs this is not an issue.

 Programs that use plenty of small functions where intermediate results

are calculated tend to use less memory overall.

 The takeaway: use functions, and plenty of them.

 If you are having trouble with memory usage contact RCS for help or take
the Python Optimization tutorial to learn more.
Tutorial Outline – Part 1

 What is Python?
 Operators
 Variables
 Functions
 Classes
 If / Else
Classes

 OOP: Object Oriented Programming

 In OOP a class is a data structure that combines data with functions that
operate on that data.

 An object is a variable whose type is a class

 Also called an instance of a class

 Classes provide a lot of power to help organize a program and can

improve your ability to re-use your own code.
 “Class Car”
Object-oriented programming
 Classes can contain data and methods
public interface
(internal functions).

 Methods can call other code inside the

class to implement complex behavior.

 This is a highly effective way of modeling

real world problems inside of a computer
program.

internal data and methods

 Object-oriented programming Track via separate variables
boston_pop = 685094
boston_sq_km = 232.1
boston_num_colleges = 35

 Python is a fully object oriented

programming (OOP) language. Boston

Population 685094

 Some familiarity with OOP is needed to Area (km2) 232.1

understand Python data structures and # of colleges 35
libraries.

 You can write your own Python classes

to define custom data types. A class lets you bundle
these into one variable
 class Student:
def __init__(self, name, buid, gpa):
Writing Your Own self.name = name
self.buid = buid
Classes self.gpa = gpa

def has_4_0(self):
return self.gpa==4.0

me = Student("RCS Instructor","U0000000",2.9)
print(me.has_4_0())

 Your own classes can be as simple or as complex as you need.

 Define your own Python classes to:
 Bundle together logically related pieces of data
 Write functions that work on specific types of data
 Improve code re-use
 Organize your code to more closely resemble the problem it is solving.
 Syntax for using Python classes
# Open a file. This returns a file object.
Create an object, which is a file = open('some_file.txt')
variable whose type is a
Python class. # Read a line from the text file.
line = file.readline()

Created by a call to the class # Get the filename

file.name # --> some_file.txt
or returned from a function.

Call a method for this object:

object_name.method_name(args…) Access internal data for this object:
object_name.data_name
Classes bundle data and functions
 In Python, calculate the area of some shapes after defining some functions.

radius = 14.0
width_square = 14.0
a1 = AreaOfCircle(radius) # ok
a2 = AreaOfSquare(width_square) # ok
a3 = AreaOfCircle(width_square) # !! OOPS

 If we defined Circle and Rectangle classes with their own area() methods…it is not
possible to miscalculate.
# create a Circle object with the radius value
c1 = Circle(radius)
r1 = Square(width_square)

a1 = c1.area()
a2 = r1.area()
When to use your own class

 A class works best when you’ve done some planning and design work
before starting your program.

 This is a topic that is best tackled after you’re comfortable with solving
programming problems with Python.

 Some tutorials on using Python classes:

W3Schools: https://www.w3schools.com/python/python_classes.asp

Python tutorial: https://docs.python.org/3.6/tutorial/classes.html

Strings in Python
 Python defines a string class – all strings in Python are objects.

 This means strings have:

 Their own internal (hidden) memory management to handle storage of the characters.

 A variety of methods (functions) accessible once you have a string object in memory.

 You can’t access string functions without a string – in Python the string
provides its own functions.
 No “strcat” / “strcmp” / etc as in C
 No “strlength” / “isletter” / etc as in Matlab
 No “nchar” / “toupper” /etc as in R
String functions
 In the Python console, create a string variable
called mystr

 type: dir(mystr) mystr = 'Hello!'

mystr.upper()

 Try out some functions: mystr.title()

mystr.isdecimal()

 Need help? Try: help(mystr.isdecimal)

help(mystr.title)
The len() function

 The len() function is not a string specific function.

 It’ll return the length of any Python object that contains any
countable thing.
len(mystr) 6

 In the case of strings it is the number of characters in the

string.
String operators
 Try using the + and += operators with strings in the
Python console.

 + concatenates strings.
 += appends strings.
 These are defined in the string class as functions that
operate on strings.

 Index strings using square brackets, starting at 0.

String operators
 Changing elements of a string by an index is not allowed:

 Python strings are immutable, i.e. they can’t be changed.

 %s means sub in
String Substitutions value
variable name
 Python provides an easy way comes after a %
to stick variable values into
strings called substitutions
Variables are listed in the
substitution order inside ()

 Syntax for one variable:

 For more than one:

 Printing: print('x: %s, y: %s, z:%s' % (xval,yval,2.0))

f-string Substitutions
 f-strings are a more name = 'Boston'
school = f'{name} University'
contemporary way to format
strings.
 Use a lowercase f before the first
quote. result = f'{mathcalc(1,2,3)}'
 Put the names of variables, or
function calls, in pairs of { } inside
the strings.
Tutorial Outline – Part 1

 What is Python?
 Operators
 Variables
 Functions
 Classes
 If / Else
If / Else

 If, elif, and else statements are used to implement conditional program
behavior
 Syntax: if Boolean_value:
…some code
elif Boolean_value:
…some other code
else:
…more code

 elif and else are not required – use them to chain together multiple
conditional statements or provide a default case.
 Try out something like this in the Spyder
editor.

 Do you get any error messages in the

console?

 Try using an elif or else statement by

itself without a preceding if. What error
message comes up?
 If / Else code blocks

 Python knows a code block has

ended when the indentation is
removed.

 Code blocks can be nested

inside others therefore if-elif-else
statements can be freely nested
within others.
 Or used in functions…
 File vs. Console Code Blocks
 Python knows a code block has ended when the indentation is removed.

 EXCEPT when typing code into the Python console. There an empty line
indicates the end of a code block.

 This sometimes causes confusion when pasting code into the console.
 Solution: try to avoid pasting more than a few lines into the Python console.