data-training

A small script enable easy training for XGBoost forecasting model on any time series data

This project implements a simple Linear Regression model from scratch and compares it to the implementation of scikit-learn, using the California Housing Dataset.

Impergrammer is a basic AI Python tool for generating text. It uses Data.txt as a style reference and attempts to rephrase any input text to match the writing style of the logs in that file. Made mostly for fun—don’t rely on it for anything serious.

Real-time facial emotion recognition system using EfficientNet, PyTorch/TensorFlow, and OpenCV, with live webcam inference and an animated robotic face UI.

A machine learning project comparing linear, ridge, and lasso regression models to forecast the US Dollar's opening price using historical data from 2003–2021. Includes full data preprocessing, training, evaluation, and model performance analysis for financial forecasting applications.

Sign language recognition application that provides learning options for the American Sign Language (ASL) while also providing transcription into text and speech. It aims to bridge communication gaps for the Deaf and Hard of Hearing (DHH) community using machine learning, specfically deep learning and computer vision.

Perbandingan 3 metode face detection berbeda (Haar Cascade, MTCNN, RetinaFace) dan face recognition dengan dataset multi-pose, multi-lighting condition.

🌐 Self-Supervised Learning AI Framework | 自监督学习AI框架