The focus of this project is to explore the domain generalisation of machine learning (ML) models which usually occurs in the presence of out-of-distribution data points. However, ML models often experience a drop in performance when applied to out-of-distribution (OOD) datasets, highlighting challenges in Domain Generalisation (DG), Domain Adaptation (DA), and Domain Shifting. This project explores these challenges by focusing on domain-specific and domain-generalised methods, vision transformers for DG, synthetic data generation for OOD analysis, and the impact of pruning on DG.