This project focuses on generating large-scale simulation datasets for studying entropy production in nonequilibrium stochastic systems. Entropy production is a key quantity for characterizing irreversibility and understanding thermodynamic processes in complex physical and biological systems. We perform extensive CPU-based simulations of stochastic dynamics to produce statistically significant trajectory data spanning a wide range of system parameters. These simulations provide the foundation for subsequent machine learning studies aimed at estimating entropy production directly from observed trajectories. The computational effort is dominated by parallel trajectory generation and data collection, requiring efficient use of high-performance computing resources to explore large parameter spaces and produce high-quality training and validation datasets. The resulting datasets will support the development and benchmarking of data-driven methods for inferring entropy production in nonequilibrium systems.