A lightweight research-oriented project analyzing how noise, missing values, and corrupted features affect machine learning model reliability. Includes synthetic data generation, multi-model training, and noise-impact evaluation with visualization. Designed as a teaching/learning artifact for data quality, robustness, and ML reliability concepts.

🎛 A comprehensive suite of DSP challenges featuring Image Restoration (Notch/Blur) and Audio Analysis (IIR/LPC). Built with Python 3.14 + uv for high-performance signal processing.

🔍 Analyze how noisy and incomplete data impacts machine learning model performance on binary classification tasks with robust evaluation methods.