clasification

Identifying Guns using Machine Learning for Image recognition with a Rapid Response System to Save Lives

An end-to-end ML platform that guarantees reproducibility across datasets, features, and models, with full lineage tracking and validation. Comes with pre-written configurations, artifact examples, comprehensive documentation, and a service that simplifies usage.

This project is designed to predict the quality of red or white wine based on various features. It utilizes Streamlit for the user interface and incorporates machine learning models from scikit-learn.

Model comparison that can classify emails as spam or not spam using machine learning algorithms.

An innovative model developed in order to detect Raga of Hindustani Classical music.

Machine learning based credit risk prediction using Random Forest, threshold optimization, and Streamlit.

This project aims to address the communication issue by developing a program capable of analyzing images from videos or providing real-time translations using images captured by a webcam. The program will analyze the speaker's hands and use a CNN model, specifically trained for this task, to translate the signs into letters and words.

Fingerprinting and classifying smartphone models based on the microphone recording characteristics, using supervised learning algorithms

Inspired by Jev and Laya, designed to go beyond classification: forecasting, representations, reinforcement learning and causal inference. Five disciplines, one ambition: master the heart of machine learning. M5PHET: Master's Five-Point Heart-Exploding Technique. A tool for real ML masters.

An end-to-end system that uses machine learning to predict anxiety probability from user responses. Supports user accounts and on-demand predictions, highlighting real-world ML deployment and backend integration.