Abstract: The increasing demand for efficient hardware acceleration in Machine Learning applications, particularly neural network inference, necessitates the development of highperformance and ...