EagerSun / DL-vs-Stat_Impute

This is a thesis project about comparing imputation performances between deep learning methods and conventional statistical methods. In this project, GAIN and VAE with One-Hot and trainable embeddings for categorical variables were built for deep learning methods. MICE and Miss-Forest were chosen for representing conventional statistical methods…
14Updated last month

Related projects

Alternatives and complementary repositories for DL-vs-Stat_Impute