Introduction

LIBBLE is the abbreviation for “a LIBrary for Big LEarning”, which provides a toolbox for big data machine learning, artificial intelligence and data science. LIBBLE contains several variants which are implemented on different parallel and distributed platforms/framework. LIBBLE-Spark is the LIBBLE variant implemented on Spark. LIBBLE-PS is the LIBBLE variant implemented with the Parameter Server framework. LIBBLE-MultiThread is the LIBBLE variant implemented on the multi-core (multi-thread) platform with a shared memory. LIBBLE-DL is the LIBBLE variant for distributed deep learning, which is implemented based on PyTorch.

LIBBLE is developed by a team under the supervision of Dr. Wu-Jun Li, an Associate Professor of LAMDA group at Nanjing University.