Circle Image

Puneet Saini

Computer Science Student

A Machine learning enthusiast and a geek, currently pursuing his degree in Bachelor of Technology in Computer Science, known by his last name — Saini, actively contributing to University's student society CESS as technical head (previously as design head), believes in promotion and contribution to Open Source Projects.

Skills

Some of the skills that I have acquired during my degree are all self-learned. They are listed below.
code

Programming Languages

33.33% Complete
30% Complete
26% Complete
Python C++ JAVA
developer_mode

Frameworks | Tools | Libraries

11% Complete
11% Complete
11% Complete
11% Complete
11% Complete
11% Complete
11% Complete
11% Complete
Pytorch 1.0 Numpy Pandas Matplotlib Keras OpenCV Git Sklearn
storage

Database

30% Complete
26% Complete
15% Complete
MySQL MS SQL Server PL/SQL
brush

Graphic designing

33.33% Complete
26% Complete
20% Complete
Adobe Photoshop Adobe Illustrator Adobe After Effects

Web Technologies

16.5% Complete
10% Complete
14% Complete
8% Complete
11% Complete
11% Complete
HTML CSS Flask ASP.NET C# VB

Education

I am currently in 7th semester of my Bachelor's degree in Technology (Computer Science) at Guru Nanak Dev University. I've always wondered about how computers work, and pursuing this degree is a step towards satisfying my curiousity.

Institute: Guru Nanak Dev University, Amritsar

Year: 2016-20

CGPA: 8.3

Institute: Bhartiya Vidya Bhavan's SL Public School, Amritsar

Year: 2015-16

Percentage: 91.80%

Institute: Bhartiya Vidya Bhavan's SL Public School, Amritsar

Year: 2013-14

CGPA: 9.0

Positions of Responsibility

I've been actively indulged with the Computer Engineering Student Society since 2017. Being a part of this community made me grow and learn the skills I lacked before. I have also been associated with School of AI, an initiative taken by Siraj Raval (Youtuber) to promote the AI culture.


Computer Engineering Students' Society

  • Volunteered in Aurum'18 Techno-Cultural Fest held in GNDU, Asr
  • Appointed as Design Head for CESS
  • Took multiple flash classes on Graphic Designing using Adobe Photoshop
  • Chief Organiser of Hack Zero to One - Hackathon in Feb 2018
  • Appointed as Technical Head for CESS
  • Delivered a workshop on Git/GitHub
  • Organised various technical workshops on C/C++ and JAVA
  • Organised CESS Summer of Code as an initiative to promote Open Source in campus
  • Organiser of Polaris'19 Techno-Cultural Fest held in GNDU, Asr
  • Chief Organiser of Hack Zero to One 2.0 - Hackathon in Mar 2019

School of AI

  • Was appointed as the dean for School of AI, Amritsar by Siraj Raval
  • Promoted and Accomodated the members on Facebook and Whatsapp
  • Had multiple meetups online with the members of School of AI Amritsar Chapter
  • Organised first meetup of School of AI Amritsar at GNDU, Asr
  • Made an agenda for next meetup and filtered out active members in group

Projects

I've hands on experience with Machine Learning in Sklearn and Pytorch and have tried implementing a variety of different machine learning algorithms. I have also deployed a project using Flask and Gunicorn on AWS. Along with these, I have a basic understanding of Linux Bash commands. I have also done ASP.NET as a part of my university curriculum and made a sample blog using ASP. You can find more projects at my GitHub profile
Github Repo for

DevToDev

DevtoDev is a blog for developers created by me and my classmate during my degree. We were given one month to learn ASP and complete the project.

Check out the project

You can check out the project at the following GitHub repository. Star it if you like it.

Github Repo for

Image Style Transfer Using CNN by Gatys

Implemented Image Style transfer with Convolution Neural Network, by Gatys using Pytorch on my own photo, extracting style from Vincent Van Gogh's Starry Night

Read the source code

Jupyter Notebook for Artistic Style Transfer.

Command Line Tool

Fast Image Style Transfer CLI

An iteration over the previous project. Optimized the style transfer time using the paper by Johnson et al. Fast Style Transfer CLI is a command line interface for training of style models used in Artistic Style transfer. It stylizes images in less than 3-5 seconds.

Check out the project

You can check out the project at the following GitHub repository. Star the repository if you liked the project. Contributors are welcomed.

Github Repo for

Flaskapp for Image Style Transfer

Implemented using Flask, and Gunicorn at Amazon Web Services (EC2 and S3). The models are built in Pytorch. This project is an extension of Image Style Transfer CLI tool.

Checkout the Website

GitHub repository and Website are linked below. Feel free to contribute to the project.

NOOPs Challenge

Hexbot Pointillism

HexBot API generates random colors(hex values). In this implementation I have tried to replicate the Pointillism effect using HexBot API and Unsplash API.

Check out the demo

Check out the demo and GitHub repo, both of them are linked below. I have wrote article on how to work with APIs which is linked below as well.

Github Repo for

Intel Scene Classification

Intel Scene Classification Challenge aims at classifying the scene depicted in an image. Participants are provided with a dataset of ~25k images from a wide range of natural scenes from all around the world.

Checkout the Demo website and notebook

The website, jupyter notebook and GitHub repo are linked below. The pretrained weights are provided in the repo. Feel free to contribute.