Over the course of this workshop, you will be introduced to various technologies.
The learning is interaction based. You will be shown how to use the different tools and how you can apply them to solve different problems.
Today's session is a little different from what is to follow. We will be showcasing a few projects that we developed using the technologies that you will learn through this bootcamp.
Since Python has a simple structure, it is very beginner friendly.
It is also perhaps why it has a large collection of libraries maintained by an ever growing community.
One such library is OpenCV, which is used for image processing.
This library was used in creating image-to-ascii, a simple command line tool that can convert images to text based art.
usage: python converter.py [-h] [-o] [-v] [-c] [-b] [-s | -w ] image_path
Convert an Image to ASCII Art
positional arguments:
image_path Path to Image File
optional arguments:
-h, --help show this help message and exit
-o , --output Path to Output File
-v, --verbose Print Verbose
-c, --color Color
-b, --braille Braille
-s , --scale Scale Factor
-w , --width New Width
Using the write command for Unix systems, we can send anyone logged in on a server a picture this way.
The following script spams a website that rewards users based on a referral program.
It creates referrals by generating random names and email ids.
# spam.py
import json
import names
import random
import string
import requests
def random_email():
''' Returns a string representing a random email '''
domains = ["hotmail.com", "gmail.com", "aol.com", "mail.com" , "mail.kz", "yahoo.com", "outlook.com"]
letters = string.ascii_lowercase[:12]
email = ''.join(random.choice(letters) for i in range(random.randint(5, 10)))
email = email + ''.join(random.choice(string.digits) for i in range(random.randint(1, 5)))
email = email + ''.join(random.choice(letters) for i in range(random.randint(1, 6)))
email = email + '@' + random.choice(domains)
return email
def spam_user():
''' Spams HOPI's website with random username '''
# Generating a random user
firstname = names.get_first_name()
lastname = names.get_last_name()
email = random_email()
# POST request parameters
url = 'https://app.viral-loops.com/api/v2/events'
headers = { 'Content-Type': 'application/json',
'Origin': 'https://behopi.com',
'Referer': 'https://behopi.com/?referralCode=Wx3CF6G&refSource=copy',
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.103 Safari/537.36',
'X-BC': '[{"key":"language","value":"en-US"},{"key":"color_depth","value":24},{"key":"pixel_ratio","value":1},{"key":"resolution","value":[1366,768]},{"key":"available_resolution","value":[1366,728]},{"key":"timezone_offset","value":-240},{"key":"session_storage","value":1},{"key":"local_storage","value":1},{"key":"indexed_db","value":1},{"key":"open_database","value":1},{"key":"cpu_class","value":"unknown"},{"key":"navigator_platform","value":"Win32"},{"key":"do_not_track","value":"unknown"}]',
'X-FP': '40e037351d7d82c641288ba002b5aa96',
'X-UCID': 'M3LqOT76h1NebJ5Pb4lvVfDDW6s' }
payload = {
"params": {
"event":"registration",
"user": {
"firstname": firstname,
"lastname": lastname,
"email": email
},
"referrer": {"referralCode":"Wx3CF6G"},
"refSource":"copy"
},
"publicToken":"M3LqOT76h1NebJ5Pb4lvVfDDW6s" }
# Sending the request
r = requests.post(url, headers=headers, data=json.dumps(payload))
print("Spammed for user {} with email {}".format(firstname, email))
for x in range(10):
spam_user()This script runs only for 10 users. This can be circumvented by using the watch Unix command.
We can use Python as a backend for websites too.
The Django Web Framework was used in creating mouseless, a website for a quiz like competition.
The concept of sorting algorithms is a good example to demonstrate the core idea behind competitive programming.
The p5.js graphic library was used to create p5-gallery to visualize these algorithms
One of the most simple projects to understand and implement is a digit recognizer, trained on the MNIST dataset.
Although Digit-Recognizer was built in JavaScript, the main model was trained using Python.
We can implement the latest state-of-the-art algorithms using Python.
The UMAP algorithm was implemented in Genomic Data Visualization Pipeline , which found patterns in genomic data for different population mappings.