ashishgupta023 Goto Github PK

ad-recommendation-via-ctrs

Developed an algorithm based on decision trees to recommend an Ad which should be shown on the website to minimize the regret based on click through rates and also handle banner blindness.

chatserver

A Chat Server serving TCP clients as well as web(HTTP) clients built using NodeJS, Express, Socket.IO, jQuery and Bootstrap.

computervision

Photometric Stereo, Image Segmentation and Stereo Vision

distributedhashtablechord

A simple distributed hash table based on chord . Implemented ring based routing, ID-space partitioning and handled node joins.

groupmessenger

A group messenger which provides TOTAL & FIFO ordering guarantees for all messages.

information-retrieval-from-news-articles

Search Engine for News Articles

machine-learning

Neural Networks, Discriminant Analysis, SVM, Logistic Regression, Linear/Non Linear Regression

multi-dimensional-news-summarization

Multi-Dimensional-News-Summarization which provides topical and chronological summarization views for a given user query/story

neo-ai-dlr

Neo-AI-DLR is a common runtime for machine learning models compiled by AWS SageMaker Neo, TVM, or TreeLite.

notebooks

real-time-eye-tracking-interface-and-gesture-recognition

The Project aims at detecting Face, Eye and Eyeballs and then track those eyeballs thus defining actions for various gestures. Apart from eyeballs, the detected face and eye have also been used to define gestures.

roboticsalgorithms

BUG2,RANSAC, Discrete Grid Bayes Localization

sagemaker-python-sdk

A library for training and deploying machine learning models on Amazon SageMaker

semantic-labeling-of-images

The supervised learning method described in this project extracts low level features such as edges, textures, RGB values, HSV values, location , number of line pixels per superpixel etc. to train the model using a Support Vector Machine and semantically label the superpixels in test set with labels such as sky, tree, road, grass, water, building, mountains & foreground objects. The results were then compared with ground truth to evaluate the accuracy of the model.

simplifiedamazondynamo

A simplified version of Amazon Dynamo which provides linearizability and availability. Implemented Quorum replication and implemented methods to handle failures , provide consistency and handle concurrency.

tvm

Open deep learning compiler stack for cpu, gpu and specialized accelerators

twittersimulator

My attempt at creating a Twitter Simulator, by considering it as two separate networks - Informational & Social. The network generated follows a Power Law Distribution

ubmlcourse

Collection of iPython notebooks for UB CSE Machine Learning course (CSE 474/574)

ud851-exercises