The Cerebros package is an ultra-precise Neural Architecture Search (NAS) / AutoML that is intended to much more closely mimic biological neurons than conventional neural network architecture strategies.

Implementation of the paper "On the Asymptotic Mean Square Error Optimality of Diffusion Models."

PyTorch implementations of the beta divergence loss.

Implementation of two new protocols in the Shuffle Model of Differential Privacy for the private summation of vector-valued messages

Our project focuses on forecasting photovoltaic (solar) power generation using a hybrid model of Gradient Boosting and LSTM. It predicts solar output with high accuracy, optimizing energy usage, improving grid stability, and enhancing renewable energy integration.

Super Resolution's the images by 3x using CNN

Practice using PyTorch include data preprocessing, linear algebra, optimization, neural networks, CNNs, and more to cover ML and DL basics

Classifying whether the credit card transaction is fraudulent or not using Logistic Regression

This repo houses a Jupyter Notebook which is intended to walk you through Gradient Descent Algorithm from scratch.

The House Price Prediction System is a comprehensive project aimed at predicting housing prices based on various attributes using advanced data analysis and machine learning techniques.

A Mathematical Intuition behind Linear Regression Algorithm

The objective is to analyze flight delays in the United States. Data from airlines, airports, and runways will be collected and processed. Machine learning models will be built using logistic regression, decision trees, and XGB classifiers. Visualizations will be created in Tableau, and Excel dashboards and SQL queries will be used for analysis.

Program for non-planar camera calibration, mean square error, RANSAC algorithm, and testing with & without noisy data using extracted 3D world and 2D image feature points.

Explains how to use ARIMA model to forecast future production units, enabling informed decision-making and planning in the electric and gas utilities sector.

This project is designed to extract sales data from a PostgreSQL database, process it, and use a Random Forest model to predict sales quantities. It also visualizes real and predicted sales for better understanding.

Comparison of common loss functions in PyTorch using MNIST dataset

This repository utilizes time series analysis to predict natural gas prices, aiding informed decisions in the energy market. Through meticulous data preprocessing, visualization, and ARIMA modeling, it provides accurate forecasts. With regression and interpolation techniques, it offers deeper insights for stakeholders, enabling proactive strategies

Implementation of different optimization algorithms. This was done as a research project for the MSc. in Computer Engineering.

Learning Project ML - Diabetes Prediction