CUDA Core Compute Libraries

Finite Field Operations on GPGPU

Written by Sem Kirkels, Nathan Bruggeman and Axel Vanherle. Grayscales an image, applies convolution, maximum pooling and minimum pooling.

Fundamentals of Accelerated Computing C/C++ is a course provided by NVIDIA.

Paperspace CORE API Documentation

Accelerating Resnet model inference using NPU.

Accelerating YOLOv3 model inference using NPU.

My solutions for NVIDIA course Fundamentals of Accelerated Computing with CUDA C/C++

Accelerating Hrnet model inference using Neural Processing Unit (NPU).

The project aims to optimize the Dynamic Time Warping (DTW) algorithm and accelerate it using Graphics Processing Units (GPUs), So that algorithm can be executed in a GPU-equipped laptop or a GPU-equipped embedded device like NVIDIA Jetson, rather than connecting to a massive server.

Talks and Presentations on Deep Learning principles,models and architectures

Advance Statistical Computing, 2019, Seoul National University