Arjun Singh's Projects
Various Data pipeline entered around Kafka, using docker images
An awesome README template to jumpstart your projects!
The goal was to perform predictive maintenance on commercial turbofan engine. The approach used here is a data-driven approach, meaning that data collected from the operational jet engine is used to perform predictive maintenance modeling. To be specific, to build a predictive model to estimate the Remaining Useful Life ( RUL) of a jet engine based on run-to-failure data of a fleet of similar jet engines. Supervised learning(Regression and Classification models) models were adapted and customized to fit our problem.
Machine learning model to predict local communities in network. The objective was to evaluate different modularity approaches on different centrality measure and build a model that do so when learning the ground truth. Tools used R; Dataset: dolphin, football, karate, pol books and Wikipedia.
A comparative analysis of supervised machine learning type, ensemble learning V/S Deep learning. The Pro's and Con's of Study are supported by the dataset complexity and computation cost and evaluation metric.
All the projects making use of deep learning frameworks
Detectron2 is FAIR's next-generation platform for object detection, segmentation and other visual recognition tasks.
End-to-End Object Detection with Transformers
The purpose of this project is to implement the learnings from PL/SQL and apply it to demo case of DEVECOR. Our objective is to build operational database do the ETL and build warehouse with DataMart’s. But for this case we replicate the framework of data warehouse with 2 databases and pushing the data after cleaning and then building the small tables needed by the respective department.
Drug Pricing prediction model, factoring in different features to learn and assess the pricing of new drugs
Use of ensemble learning algorithm to improve the accuracy of Decision tree model. Build a predictive model to understand key parameters affecting USA income.
An analysis of prospective and current student data. The aim for this project is to constitute a geographical database of EISTI’s students with their personal address, campus address and the addresses where their internships took place. Having that, some statistics will e performed to calculate the distance between: « home – campus », « home – internships location » and « campus – internships location » respectively. These statistics will serve to know the search trends and to generate a classification in order to help users in charge of business relations.
Image Classification problem, Cats v/s Dogs Model. Browse to https://imgclassification.herokuapp.com/ for the deployment via Heroku
The project was to build an Artificial neural network for OCR but the underlying objective was to study the feature extraction and to understand the behavior(signals) of the characters extracted from image. Image processing has high dimension and how to handle and process them was the second objective. Tools used Matlab to simulate the model and Mathcad for signal processing and building logics; Dataset used: alphanumeric text.
Kats, a kit to analyze time series data, a lightweight, easy-to-use, generalizable, and extendable framework to perform time series analysis, from understanding the key statistics and characteristics, detecting change points and anomalies, to forecasting future trends.
Long Text research for BERT transformers
Classification of different music genre
this is my description
The goal of the project was to build an approximation algorithm for finding the house(rectangle) with largest surface area inside a plot of land(non-convex polygon). Exploration of solution using two heuristic's approach ----- 1.Taboo Search ------ 2.Particle Swarm Optimization
Stochastic_and_deterministic algorithms for Rosenbrock And Himmelblau. A Study of different algorithm and evaluating performance on a grid