H2O is an Open Source, Distributed, Fast & Scalable Machine Learning Platform: Deep Learning, Gradient Boosting (GBM) & XGBoost, Random Forest, Generalized Linear Modeling (GLM with Elastic Net), K-Means, PCA, Generalized Additive Models (GAM), RuleFit, Support Vector Machine (SVM), Stacked Ensembles, Automatic Machine Learning (AutoML), etc.

Fast and Accurate ML in 3 Lines of Code

Detection of Human Edited Images using CNN, VGG16, Xception, ELA, Ensemble Learning.

🏆2nd solution in web ad CTR predict competition🏆

Dataset of coral images conditions in Koh Tao, Thailand

A collection of 8 Applied Data Science projects.

A face recognition model build with an ensemble of popular pre-trained models like FaceNet and OpenFace, on training with a dataset of 31 celebrity images. Built an application which can recognise a new person based on stored embedding of him and relate his facial features to the 31 celebrities it was trained.

Build a Web App called AI-Powered Heart Disease Risk Assessment App

A Julia machine learning framework

The solutions of the assignments and projects of the course CS412: Machine Learning (Sabanci University).

Feature Engineering, Regression, Classification, Model Explanation. My 2 biggest projects exploring the link between economic indicators and U.S. presidential election results.

A Streamlit web application for classifying plant diseases using deep learning models.

Ensemble based approach compared to traditional machine learning models

This model utilizes regression models and accurately predicts employee salaries based on experience, previous CTC, and job roles, promoting fair salary structures and optimizing resource allocation for streamlined HR operations.

This project employs ensemble learning methods to forecast cybercrime rates, utilizing datasets with population, internet subscriptions, and crime incidents. By analyzing trends and employing metrics like R2 Score and Mean Squared Error, it aims to enhance prediction accuracy and provide insights for effective prevention strategies.

Merlion: A Machine Learning Framework for Time Series Intelligence

To develop an advance forecasting model that adeptly incorporates solar irradiance data, leveraging its predictive capabilities to elevate forecasting performance and reliability.

This code performs email spam classification using three machine learning models: Naive Bayes, Support Vector Machines (SVM), and Random Forest Classifier. It evaluates their performance using accuracy scores and classification reports, ultimately identifying Random Forest Classifier as the best performer among the three.

This project uses an ensemble of CNN, RNN, and VGG16 models to enhance CIFAR-10 image classification accuracy and robustness. By combining multiple architectures, we significantly outperform single-model approaches, achieving superior classification performance.