modeloptimization

Machine Learning model optimization and validation for Nutri-Score prediction using Random Forest, hyperparameter tuning, cross-validation, and hold-out evaluation.

🔹 Hands-on experience in building and training ML & DL models using TensorFlow 🤖 🔹 Skilled with Keras API, Tensors, & Computational Graphs ⚙️ 🔹 Developed projects like Image Classification & Neural Networks 🧠 🔹 Strong understanding of data preprocessing model evaluation 🔹 Exploring model

SafeGuard-CV is a real-time computer vision system that detects missing PPE (helmets) on construction sites using a fine-tuned YOLOv8s model, optimized for edge deployment via ONNX FP16 quantization.

A comprehensive comparative study of 10+ feature selection techniques (including RFE and SHAP) to optimize ML models. Achieved a 73% reduction in feature space while maintaining >96% accuracy, highlighting key trade-offs between performance efficiency and model interpretability for production environments.