Fitness Tracker Based on ML

Imagine yourself at the gym, pushing through your strength training routine, but wouldn’t it be amazing if you could ditch the expensive personal trainer and instead have a smart companion right on your wrist? Picture a device that not only tracks every move you make during your weightlifting sessions but also counts your repetitions, ensuring you stay on track with your fitness goals. That’s precisely the problem we’re tackling: the lack of automated tracking systems for free weight exercises, particularly in the realm of strength training.

Features

Automated Exercise Tracking: Tracks various barbell exercises during strength training sessions, eliminating the need for expensive personal trainers.

Repetition Counting: Accurately counts repetitions, helping users stay on track with their fitness goals.

Exercise Detection: Identifies different exercises being performed using accelerometer and gyroscope data.

Machine Learning Integration: Utilizes advanced machine learning models built through a comprehensive lifecycle, including data cleaning, preprocessing, feature engineering, model selection, and evaluation.

Superior Performance: Achieves better performance compared to existing solutions, such as the work done by Dave Ebbelaar in his master's degree project.

Deployment and MLOps: Implements modern deployment strategies using Flask and MLOps tools like MLflow, along with continuous integration and continuous delivery (CI/CD) practices.