Mars is a tensor-based unified framework for large-scale data computation which scales numpy, pandas, scikit-learn and Python functions.

Framework for correlating two or more well logs using feature vectors generated from CNN's in Pytorch

Joblib-like interface for parallel GPU computations (e.g. data preprocessing)

A Proximal Policy Optimization Approach to Detect Spoofing in Algorithmic Trading

A machine learning project to predict smoking status (Smoker/Non-Smoker) using health and lifestyle data. It includes data preprocessing, model training, evaluation, visualizations, and FastAPI-based deployment, supporting CI/CD and multiple datasets for robustness.

This repository contains two machine learning projects: a Streamlit-based app for predicting household power consumption using a Random Forest model, and a Tkinter GUI application for predicting telecom customer churn using LightGBM with SMOTE handling.

An efficient text classification pipeline for email subjects, leveraging NLP techniques and Multinomial Naive Bayes. Easily preprocess data, train the model, and categorize new email subjects. Ideal for NLP enthusiasts and those building practical email categorization systems using Python.

A step-by-step guide to master various aspects of Joblib for parallel computing in Python

🚀 A flexible Python library for easy handling and conversion of Hierarchical, Tabular, and Serialized data formats.

A GitHub WebCrawler

Predicts which telecom customers are likely to churn with 95% accuracy using real-world data features from usage, billing, and support data. Implements Sturges-based binning, one-hot encoding, stratified 80/20 train-test split, and a two-level ensemble pipeline with soft voting. Achieves 94.60% accuracy, 0.8968 AUC, 0.8675 precision, 0.7423 recall.

An IA model that detects whether a given verse is from the Bible or not

A basic malware detector using Machine Learning

Python scripts that scrape US gas prices

PyPOLAR is a Python-based app for analyzing polarization-resolved microscopy data to measure molecular orientation and order in biological samples

A bot designed to answer live trivia game questions.

magic-wormholing pickled objects made simple

StockSage is a production-ready RESTful API built with FastAPI and Docker to serve machine learning models. It allows users to send stock data and receive price predictions using the Prophet model. Designed for scalability, it follows best practices in API deployment, containerization, and production model serving.

Using Protocol Buffers and gRPC client-server communication to deploy a scikit-learn joblib exported model.

Detects whether a message is SPAM or NOT SPAM using Multinomial Naive Bayes and TF-IDF vectorization. Built with Streamlit for real-time prediction.