Coding Club Documentation

E-Rickshaw Tracking App

We made an E-Rickshaw Tracking system with login for driver, live tracking of vehicles for user. Firebase was used to maintain the authentication and storage of live location. Firebase Cloud Store for recording the complaints.
Link to the app:

Download Now

DUBG

DUBG is a mixed reality app for efficient post disaster management. Rescuer teams are very busy during operation, which is why we built our app based on Mixed Reality so that the rescuers don’t waste time on planning their rescue drive and operating any device for communication. Our vision is to create an Augmented Reality (AR) based navigation system which disaster rescue team can use in a hassle free manner. The features of our application can also be extended to help army to communicate and coordinate efficaciously in unknown terrains.

Study Portal

Study Portal is a Django based web application which aims to serve as a study resource catalog and a discussion forum.
Study Resource Catalog - We have implemented a system where people can easily access study materials (present in well-organized filesystems) of various courses. Anyone can also upload these resources which will then be added to the catalog depending on whether they are approved by the corresponding course’s admin.
Discussion Forum - Any user of the application can post to the forum, which is quite similar to any other social media with features like upvotes, downvotes, comments, share etc. People can post their queries and get their answers using this facility

IMDB Rating Predictor

The aim of the project is to analyse the sentiment of a user comment on an IMDB rated movie, and predict its rating using deep learning and natural language processing methods. The rating, which is an integer value from 0 to 10, is predicted as a multi-class classification task.

The sentiments depict the overview of the wider public opinion about a movie, and thus can help in predicting the box office revenue earned by that movie. Another aspect of sentiment analysis is that it includes rich structured information about the individuals involved in the process, which can help in identifying the preferences of an individual. This project was a derivative from the work done for Intrusion Detection with ML project.

Trex-Rush

In this project we play Google's T-rex game using Reinforcement Learning. The RL algorithm is based on the Deep Q-Learning algorithm. This demonstrates how a computer learns to play just by looking at the screen pixels and receiving a reward when the game score is increased. This gives a good essence of reinforcement learning.

SAIL Yearbook Portal

Yearbook portal is a Django based web application created for the graduating batch. The portal is similar to slam book where users can write testimonials about each other. Users can login with IITG login credentials.

Main features

• Testimonials - Any user can write testimonials about another user, few of which get printed in the final copy of yearbook of graduating batch.
• Polls - One can create a poll for which graduating batch nominate candidates and the winner is decided.
• Profiles - Each user has a profile in which he can answer questions about his experiences at IITG.

Face Recognition System

This a Deep learning based face recognition system for security purposes. A fully automated system was designed to capture the video frame from webcam , preprocess it to extract the face images from video frame and then feed it to a deep convolutional neural network to generate the 128- dimensional face encoding.This encoding is compared with those in the database using Euclidean distance measure and if the distance goes above a threshold, it is claimed to be unidentified.

This uses FaceNet, a specialized deep learning algorithm for face recognition

Auto Gameplay (Pommerman Competition)

Pommerman is stylistically similar to Bomberman, the famous game from Nintendo. Every battle starts on a randomly drawn symmetric 11x11 grid. There are four agents, one in each corner. If applicable, the agent's teammate will be on the kitty corner.

We will be participating in “Team” which is the second competition which will be held live at NIPS 2018. Entrants must be submitted by 11:59pm Est Nov 21st.

The main aim of the competition is to develop an agent which plays the Pommerman game with a high success rate. We have developed a similar project in the summer which plays any Atari based game. Its development was majorly hindered due to the low compute power available to us. We do not want this to stop us from competing in this competition. That's the reason we applied for a budget of ₹10,000 for AWS credits, which gives us 150hrs of training time on p2x.large instance, which will be sufficient for us to train our agent on.

Prizes for the competition are as follows:
1st $4k USD - $6k GCE credit
2nd $2k USD - $4k GCE credit
3rd $1k USD - $2k GCE credit
4th $2k GCE credit

Coding Club Website v2.0

Coding Club is back with a new and improved Django framework website with enhanced and more secure features than the previous PHP based website.

The Website organizes activities related to Development(Web Development, Open Source, Game/App development), programming contests, hackathons, et al.

*. Back-end programming is coded in python and database stored using mysqlite using the Django framework.
*. Front-end programming is coded in JavaScript making it dynamic and bilateral.
*. The use of materialized CSS and Bootstrap makes the portal interactive and responsive.
*. The use of the Django User Authentication model for an extensive Login system makes handling ongoing and completed projects dynamic for the admin and abstract for the user at the same time.
*. The site is up to date with all the contact info of Coding Club Members 2019.