Accelerate the Bridging of ML and DL with NVIDIA-Accelerated Apache MXNet 2.0

Deep Learning (DL) community has largely evolved independently from the large NumPy community of data science (DS) and machine learning (ML) community. While most DL frameworks now provide NumPy-like array library, they differ in the definition of the operations which creates a steeper learning curve of DL for ML and DS practitioners. The next major version, 2.0, of Apache MXNet (incubating) seeks to bridge the fragmented DL and ML ecosystem. It provides NumPy-compatible programming experience and Gluon 2.0 interface to enable NumPy for DL, and also enables GPU acceleration, auto-differentiation, and high-performance one-click deployment to the NumPy ecosystem. From training DL models at scale on thousands of GPUs to inferencing models on small Jetson edge devices, MXNet does it well with the use of the latest GPU features and exciting performance advancements such as Run-time Compilation (RTC), fusion and faster dependency engine scheduler.