Basic Python

ML4HEP 2026 Pre-school Lectures

Subir Sarkar, SINP

Lecture 1, 11/05/2026
 Course Outline
● Python in a Nutshell
● Python interactive
● Basic data types
● Variable scope / global vs local scope
● Type system & conversion
● Control flow, loop
● Basic I/O

2
Python's place in the computing world

3
 Programming Languages in the Scientific World
Compiled Byte Compiled Interpreted
(partly compiled + interpreted)

Fortran Tcl
cess native libraries)
Extension API (ac
C
m be d d in g inside C/C++
E
C++

● dynamic type
● portable
● static type
● relatively slow
● fast
● scripting/embedding
● source level portability
● interface to C/C++/Fortran 4
 Programming Languages in the Scientific World
Compiled Byte Compiled Interpreted
(partly compiled + interpreted)

Fortran Tcl
cess native libraries)
Extension API (ac
C Python byte compiled
m be d d in g inside C/C++
E
C++

● dynamic type
● portable
● static type
● relatively slow
● fast
● scripting/embedding
● source level portability
● interface to C/C++/Fortran 5
 Programming Languages in the Scientific World
Compiled Byte Compiled Interpreted
(partly compiled + interpreted)

Fortran Tcl
cess native libraries)
Exte nsion API (ac
C Python byte compiled
m be d d in g inside C/C++
Java E
Nat
C++ ive
Inte
rfac
e Java
● dynamic type
● portable
● static type ● static type ● relatively slow
● fast ● secure ● scripting/embedding
● source level portability ● portable ● interface to C/C++/Fortran/Java 6
 Programming Languages in the Scientific World
Compiled Byte Compiled Interpreted
(partly compiled + interpreted)

Fortran Tcl
cess native libraries)
Exte nsion API (ac
C Python byte compiled
m be d d in g inside C/C++
Java E
Nat
C++ ive
Inte
rfac
Java
BeanShell
e Groovy
Jython
● dynamic type
● portable
● static type ● static type ● relatively slow
● fast ● secure ● scripting/embedding
● source level portability ● portable ● interface to C/C++/Fortran/Java 7
 Programming Languages in the Scientific World
Compiled Byte Compiled Interpreted
(partly compiled + interpreted)

Fortran Tcl
cess native libraries)
Exte nsion API (ac
C Python byte compiled
m be d d in g inside C/C++
E
Java
Nat R
C++ ive
Inte
rfac Perl, PHP,
Java
BeanShell
e Groovy JavaScript
Jython
● dynamic type
● portable
● static type ● static type ● relatively slow
● fast ● secure ● scripting/embedding
● source level portability ● portable ● interface to C/C++/Fortran/Java 8
 Programming Languages in the Scientific World
Compiled Byte Compiled Interpreted
(partly compiled + interpreted)

Fortran Tcl
cess native libraries)
Exte nsion API (ac
C Python byte compiled
m be d d in g inside C/C++
E
Go, Rust Java
Nat R
C++ ive
Inte
rfac Perl, PHP,
Java
BeanShell
Julia (JIT compiled) e Groovy JavaScript
Jython
● dynamic type
● portable
● static type ● static type ● relatively slow
● fast ● secure ● scripting/embedding
● source level portability ● portable ● interface to C/C++/Fortran/Java 9
 Popularity of Python

TIOBE ratings are based on the number of skilled engineers world-wide, courses and third party vendors. Popular
web sites Google, Amazon, Wikipedia, Bing and more than 20 others are used to calculate the ratings .

10
 Popularity of Python
[Link]

The PYPL PopularitY of Programming Language Index is created by analyzing how often language
tutorials are searched on Google.

Python has slowly become the numero uno in the last 5 years, thanks to its ubiquitous role in Machine Learning!
11
Python at a glance

12
 Python Design Highlights
● Supports all major programming
paradigms
○ procedural, object oriented and
functional
○ imperative as well as declarative
● Simple syntax, white space
significant, pass by-reference
● Dynamically typed, loosely bound
● Scripting & driver language
○ binds different components together
○ seamlessly supports multi-language
[Link]
environment (Extension &
Embedding)
13
 Python Design Highlights
● One can programmatically build Python code as string
on-the-fly, and execute inside a running Python program
● Object Orientation from ground up (no data hiding, though!)
● A very high level language (VHLL)
○ built-in support for exception
○ introspection is an integral part of the language
● High level API to seamlessly access C/C++/Java/Fortran libraries
○ opens up infinite possibilities
○ Python objects can be directly created from the native library (.so)

14
 Integrated Development Environment
● Interactive environment
○ python shell
○ ipython ([Link]
○ jupyter notebook - [Link]
○ IDLE, spyder, PyCharm
○ SWAN - [Link] - a platform 15 to

perform interactive data analysis on the cloud

 Python application domain
● Scientific computing, data science
○ NumPy, pandas, PyTables, PyROOT,
Machine Learning frameworks (classical,
quantum)
● System administration
● Web frameworks
○ CGI, CherryPy, Flask, Django,
TurboGears and a hell lot more
● Interface to databases, XML and json
parsers
● GUI development
○ Tkinter, PyGtk, PyQT/PySide, wxPython

[Link]
16
 Python as an interpreted language
● A Python interpreter source code
○ parses a Python
program/source file (.py)
external
■ compiles into byte-code parse
libraries
(.pyc)
○ executes byte-code edit
■ interprets on a virtual byte string of bytes
machine (VM) compile
○ no need for compilation test
and linking phases
■ no binary executable interpret on the VM
run

[Link]
17
 Portability & Performance
● Python source code is inherently portable across
platforms/architectures
○ the underlying virtual machine (VM) instance takes care of
architecture dependence
○ in most cases you pay a performance penalty
■ vis-a-vis Fortran/C/C++/Go/Rust(/Julia)
■ benefits usually outweigh lack of performance
■ there are alternative implementations of Python to address
performance issues with certain limitations
● Cython - uses additional C/C++ extension libraries
● PyPy (w/ JIT) - works with pure Python, not with C extension
● numba (w/ JIT) - sprinkle decorators to accelerate code execution 18
 Portability & Performance
Fast
Fortran C Rust Language Energy Time Memory
C++ C 1.00 1.00 1.17

Julia Rust 1.03 1.04 1.54

Java
C++ 1.34 1.56 1.34
Easy
Fortran 2.52 1.89 1.24
Hard
Python Java 1.95 4.20 6.01
R Julia na na na

Python 75.88 71.90 2.80

R na na na
Slow
[Link] 19
Interactive Python
20
 Interactive Python
$ python3
>>> credits
>>> copyright
>>> help(1) # help on integer Get information,
>>> help() extensive help
help> keywords
help> quit
>>>

There are more than one ways to exit an interactive session

>>> CTRL-D
>>> quit()
>>> raise SystemExit
>>> import sys
>>> [Link]() 21
 Python as a calculator
$ python3
Python 3.8.10 (default, Mar 15 2022, 12:22:08)
[GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> 2 + 2
4
>>> a = 2
>>> b = 2
>>> c = a * b
>>> print(c)
4
>>> bin(16)
'0b10000'
>>> pow(2,3) Python does not load the full math
8 library in memory
>>> log(10)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
NameError: name 'log' is not defined

● Two major releases in use: python2 and python3

● python3 is the default on modern OSes, breaks backward compatibility
● python3 is used in this course 22
 Python as a calculator
>>> import math # load a module
>>> dir(math) # see what it offers
[... 'log', 'log10', 'log1p', 'modf', 'pi', 'pow',
'radians', 'sin', 'sinh', 'sqrt', 'tan', 'tanh',
'trunc']
>>> [Link]
3.141592653589793
>>> from math import sqrt,pow
>>> pow(2,3)
8.0
>>> sqrt(10)
3.1622776601683795
>>> _ # result of the last expression
3.1622776601683795 23
Code Execution: script, eval, exec

24
 Python script (on Linux)
Shell magic!
#!/usr/bin/env python3
# [Link] Find the default
for i in list(range(10)): Python interpreter
print(i, end=" ")
[1]$ python [Link]
0 1 2 3 4 5 6 7 8 9
[2]$ chmod a+x [Link]
[3]$ ./[Link]
Turn on execution bit
0 1 2 3 4 5 6 7 8 9
25
 eval
$ python3
>>> result = eval('1.0 + 1.0')
>>> print(result, type(result))
2.0 <class 'float'>
>>> print(eval("len('hello world!')"))
12
>>> print(eval("__import__('os').getcwd()"))
/home/sarkar/python/examples

26
 exec
● one can construct Python code as string and execute
on-the-fly from within a running Python program
>>> code = 'print("hello world!")'
>>> exec(code)
hello world!

27
Data type, variable scope, namespace

28
 Built-in Data Types
Data type Example
Numbers 3.1415, 1234, 9999999999, 3+4j

String “Hello”, “guido v”

List [1, [2,'three'], 4]

{'compiled': 'Fortran,C,C++',
Dictionary
'interpreted': 'Perl, Python, Ruby'}
Tuple 1,'world!',3,4 or (1,'world!', 2, 3)
f = open('[Link]',
File
'r').readlines()

29
 Python - dynamically typed, loosely bound
● Name (refer) a variable, no need to declare the type
○ variable type is context sensitive
○ from Python 3.6 static type is allowed, basically as a hint
● Python keeps track of type of an object referenced by a variable name during it's lifetime i.e
Python is NOT weakly typed

30
 Python - dynamically typed, loosely bound
● Name (refer) a variable, no need to declare the type
○ variable type is context sensitive
○ from Python 3.6 static type is allowed, basically as a hint
● Python keeps track of type of an object referenced by a variable name during it's lifetime i.e
Python is NOT weakly typed

>>> a = False
>>> print(type(a))
<class 'bool'>

31
 Python - dynamically typed, loosely bound
● Name (refer) a variable, no need to declare the type
○ variable type is context sensitive
○ from Python 3.6 static type is allowed, basically as a hint
● Python keeps track of type of an object referenced by a variable name during it's lifetime i.e
Python is NOT weakly typed

>>> a = False
>>> print(type(a))
<class 'bool'>
>>> a = 5
>>> print(type(a))
<class 'int'>

32
 Python - dynamically typed, loosely bound
● Name (refer) a variable, no need to declare the type
○ variable type is context sensitive
○ from Python 3.6 static type is allowed, basically as a hint
● Python keeps track of type of an object referenced by a variable name during it's lifetime i.e
Python is NOT weakly typed

>>> a = False
>>> print(type(a))
<class 'bool'>
>>> a = 5
>>> print(type(a))
<class 'int'>
>>> a = "Hello World!"
>>> print(type(a))
<class 'str'>
33
 Python - dynamically typed, loosely bound
● Name (refer) a variable, no need to declare the type
○ variable type is context sensitive
○ from Python 3.6 static type is allowed, basically as a hint
● Python keeps track of type of an object referenced by a variable name during it's lifetime i.e
Python is NOT weakly typed

>>> a = False >>> a = [1,2,3,4]

>>> print(type(a)) >>> print(type(a))
<class 'bool'> <class 'list'>
>>> a = 5
>>> print(type(a))
<class 'int'>
>>> a = "Hello World!"
>>> print(type(a))
<class 'str'>
34
 Python - dynamically typed, loosely bound
● Name (refer) a variable, no need to declare the type
○ variable type is context sensitive
○ from Python 3.6 static type is allowed, basically as a hint
● Python keeps track of type of an object referenced by a variable name during it's lifetime i.e
Python is NOT weakly typed

>>> a = False >>> a = [1,2,3,4]

>>> print(type(a)) >>> print(type(a))
<class 'bool'> <class 'list'>
>>> a = 5 >>> a = 1,2,3,4
>>> print(type(a)) >>> print(type(a))
<class 'int'> <class 'tuple'>
>>> a = "Hello World!"
>>> print(type(a))
<class 'str'>
35
 Python - dynamically typed, loosely bound
● Name (refer) a variable, no need to declare the type
○ variable type is context sensitive
○ from Python 3.6 static type is allowed, basically as a hint
● Python keeps track of type of an object referenced by a variable name during it's lifetime i.e
Python is NOT weakly typed

>>> a = False >>> a = [1,2,3,4]

>>> print(type(a)) >>> print(type(a))
<class 'bool'> <class 'list'>
>>> a = 5 >>> a = 1,2,3,4
>>> print(type(a)) >>> print(type(a))
<class 'int'> <class 'tuple'>
>>> a = "Hello World!" >>> a = None
>>> print(type(a)) >>> print(type(a))
<class 'str'> <class 'NoneType'>
36
 Variable Scope & Lifetime
● Scope
○ part of a program where a variable is accessible
● Lifetime
○ duration for which a variable is accessible
● In general, variables defined
○ in the main body of a program have global scope
■ visible throughout the file, and also inside any file which
imports that file
■ global scope is usually a bad idea
○ inside a function have local scope
○ in a class have local scope, but can be accessed from outside
through the object (no data hiding)
37
 Variable Scope & Lifetime
# global # local
a = 0 def test_scope(c):
d = 3
if a == 0: print(c, d)
b = 1
# call the function
# what do we expect? d = 5
print(b) test_scope(7) 38

# what do we expect?
print(c, d)
 Type System
● static vs dynamic
○ type checking at compile time vs leaving the responsibility to runtime
● strong vs weak
○ how runtime treats types
○ 1 + “1” gives error for strongly typed language (e.g Python)
● explicit vs implicit
○ how type conversion takes place

>>> int("Hello")
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ValueError: invalid literal for int() with base 10: 'Hello'
>>> int('5')
5
>>> float('5.2')
5.2 39
 Datatype conversion
int(x [,base]) - converts x to an integer. Base specifies the base if x
is a string
long(x [,base]) - converts x to a long integer. Base specifies the base
if x is a string
float(x) - converts x to a floating-point number
complex(real [,imag]) - creates a complex number
str(x) - converts object x to a string representation
repr(x) - converts object x to an expression string
eval(str) - evaluates a string and returns an object
tuple(s) - converts s to a tuple
list(s) - converts s to a list
set(s) - converts s to a set
dict(d) - creates a dictionary. d must be a sequence of (key,value)
tuples
chr(x) - converts an integer to a character
ord(x) - converts a single character to its integer value
hex(x) - converts an integer to a hexadecimal string
oct(x) - converts an integer to an octal string
40
 True or False
● The empty string is mapped to False, every other string is mapped to True
>>> s = ''
>>> if s:
... print('hello')
...
>>> s = 'rob'
>>> if s:
... print('hello')
...
hello
● For integers, 0 is mapped to False and every other value to True
● For floating point numbers, 0.0 is mapped to False and every other value to True
>>> a = 1
>>> if a:
... print('hello')
...
hello 41
 Control Flow
x = int(input("Enter an integer: "))
if x < 0:
x = 0 standard input
print('Negative changed to zero')
elif x == 0:
print('Zero')
elif x == 1:
print('One')
else:
print('More than one')

● Note
○ if-elif-else construct slightly unconventional
○ indentation is mandatory 42
 pass statements
The pass statement does nothing. It can be used when a statement
is required syntactically but the program requires no action

>>> while True: # infinite loop

... pass
...
i = int(input("Enter an integer (>=0): "))
if i > 0:
pass # place-holder for future code
else:
print("i is zero")

43
 Loop: for & while
>>> list(range(10))
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
>>> for i in range(10): for
... print(i, end = " ")
...
0 1 2 3 4 5 6 7 8 9

44
 Loop: for & while
>>> list(range(10))
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
>>> for i in range(10): for
... print(i, end = " ")
...
0 1 2 3 4 5 6 7 8 9

>>> index = 0 while

>>> while index < 10:
... print(index, end = " ")
... index += 1 45
 Loop - continue & break
for i in range(10):
if i%2 == 0: continue
print(i, end = " ")

46
 Loop - continue & break
for i in range(10):
if i%2 == 0: continue
print(i, end = " ")

for i in range(10):
if i**2 > 25: break
print(i, end = " ")

47
 pass by reference
def square(items):
for i,x in enumerate(items):
items[i] = x * x # modify items in-place

a = [1, 2, 3, 4, 5]
print(a)
square(a) # changes a to [1, 4, 9, 16, 25]
print(a)

48
 Iteration over collection
# list
for element in [1, 2, 3]:
print(element, end = " ")

49
 Iteration over collection
# list
for element in [1, 2, 3]:
print(element, end = " ")

# tuple
for element in (1, 2, 3):
print(element, end = " ")

50
 Iteration over collection
# list
for element in [1, 2, 3]:
print(element, end = " ")

# tuple
for element in (1, 2, 3):
print(element, end = " ")

# dictionary
for key in {'one':1,'two':2}:
print(key, end = " ")

51
 Iteration over collection
# list # character string
for element in [1, 2, 3]: for char in "123":
print(element, end = " ") print(char, end = " ")

# tuple
for element in (1, 2, 3):
print(element, end = " ")

# dictionary
for key in {'one':1,'two':2}:
print(key, end = " ")

52
 Iteration over collection
# list # character string
for element in [1, 2, 3]: for char in "123":
print(element, end = " ") print(char, end = " ")

# tuple # file
for element in (1, 2, 3): for line in open("[Link]"):
print(element, end = " ") print(line, end = " ")

# dictionary
for key in {'one':1,'two':2}:
print(key, end = " ")

53