I'm Mridul, an AI/ML enthusiast with hands-on experience in OpenCV, Flask, scikit-learn, and TensorFlow. I enjoy building practical and innovative AI solutions, transforming ideas into impactful real-world applications.
Download CV
NumPy
Pandas
TensorFlow
OpenCV
YOLO
Matplotlib
Seaborn
Power BI
Excel
Scikit-learn
XGBoost
LightGBM
Flask
Welcome to FIT-U, an AI-powered fitness application designed to make your workouts smarter, easier, and more effective. Hosted on Render, this app leverages computer vision and pose estimation to track your exercises in real-time.
This project uses YOLOv8 for object detection combined with OpenCV for video processing and SORT to track vehicles in real-time. The goal is to detect and count vehicles in a video feed.
This project focused on predicting NYC taxi fares using XGBoost, successfully reducing the RMSE to $3.22. Enhanced computational efficiency while maintaining high prediction accuracy.
Predicting whether a mushroom is edible or poisonous based on physical characteristics. XGBoost achieved an exceptional MCC of 98.3%, showcasing its effectiveness in classification.