Hi! I am Rohith Reddy.
- Websites Hosted on AWS:
- Library Management System
- Movie Review/Rating Website(CINE-SHOW)
Python app/service which decodes sequences of digits from natural images (live camera). Built on Convolutional Neural Networks (CNN) and Recurrent Neural Networks(RNN) using Python-Tensorflow framework.
Object classificaiton on CIFAR-10 dataset. Dataset contains 60000 32x32 images. Classifiers used are Softmax Classifer and 2-Layer NNs and programmed using Python-Tensorflow.
Implemented a database engine based on a hybrid between MySQL and SQLite. Only supports actions on a single table at a time. Approach similar to InnoDB engine. Developed purely in Java
Predicted the functionality of the water pump given a set of test data. Implemented the project using Ensemble methods: Gradient Boosting and Random Forest. Team’s current rank stands at 137 out of 2659 competitors. The application is developed in R using multiple packages.
Implementing a snapshot protocol to ensure a consistent global state in a multi process network. These snapshot or recovery points are later used to restore the system to a valid state in case of a system failure. Developed in Java using socket programming.
Implementing the two protocols which enforce Mutual Exclusion in a multi process network. The first one is Lamport Mutex Algorithm and the second is Ricart-Agarwala Mutex algorithm. Ricart-Agarwala is the faster and less overhead of the two. Developed in Java using socket programming.
Concepts include Harris Corners, Lucas-Kanade, SIFT, RANSAC, Gaussian & Laplacian Pyramids, Particle Filter Tracking. All of the concepts are implemented using Python with the help of OpenCV library.
Developed a scalable web application to review/rate wide range of movies. Users can rate, review and add favorite movies to their list. Project is built using MongoDB, AngularJS, ExpressJS and NodeJS (MEAN Stack).
Developed a web application to manage library books inventory. Created database schema and populated the database from raw data. Some of the features include searching the catalog, checking in and checking out books etc. GUI is built using PHP and database maintained using MySQL.
Implemented image based cryptographic technique known as 'Visual Cryptography'. Implemted on GPU and Multi-core processors. Developed in CUDA programming language which is the C language extension for GPU/Multi-core programming.
Implemented the infamous Netflix prize competition. Predict the rating a user might give for any given movie. Developed in python without using any other pre-written packages and achived RMSE of 0.97.
Implemented a file system from scratch based on the UNIX V6 file system. Supports all major file system functions such as create new directory, copy files, list files etc. Project is implemented using C language.