Example of Gini calculation

Example of Gini calculation

This repository serves as the official code release of the paper "REMOT: A Hardware-Software Architecture for Attention-Guided Multi-Object Tracking with Dynamic Vision Sensors on FPGAs" (FPGA'22).
REMOT is a hardware/software architecture for Multi-Object Tracking using Dynamic Vision Sensors on FPGA. It's designed around the concept of an attention unit (AU). Each AU will only pay attention to a specific region of interest, which is designed to changed as the object moves. In REMOT, a layer...

This repo serves as the official implementation of the paper "DPACS: Hardware Accelerated Dynamic Neural Network Pruning through Algorithm-Architecture Co-design".
DPCAS is an algorithm-architecture co-design framework for dynamic neural network pruning. It utilizes a hardware-aware dynamic spatial and channel pruning mechanism in conjunction with a dynamic dataflow engine in hardware to facilitate efficient processing of the pruned network.
The code is also avaliable on github: https://github.com/CASR-HKU/DPACS
Our group page: https://casr.eee.hku.hk/publication/dpacs-asplos23/

Estimation of submarine groundwater discharge (SGD) has traditionally relied on isotopic mass balance calculation, which is a form of regional box modeling. Here we propose an entirely new approach for SGD estimation by coupling a general marine hydrodynamic simulation model with a global optimization algorithm. DYCORS (DYnamic COordinate search using Response Surface models) is a surrogate-based optimization approach for high-dimensional, expensive, and black-box function’s optimization (Regis & Shoemaker, 2013). Because of its effectiveness and adaptability,...

This repository serves as the official code release of the paper "REMOT: A Hardware-Software Architecture for Attention-Guided Multi-Object Tracking with Dynamic Vision Sensors on FPGAs" (FPGA'22).
REMOT is a hardware/software architecture for Multi-Object Tracking using Dynamic Vision Sensors on FPGA. It's designed around the concept of an attention unit (AU). Each AU will only pay attention to a specific region of interest, which is designed to changed as the object moves. In REMOT, a layer...

DDFS3D is a set of open-source code which utilizes DDM (displacement discontinuity method), FSM (fictitious stress method) and the hybrid of DD-FS to efficiently and accurately simulate fractures embedded in large 3D domains. DDFS3D is developed in Fortran language with a number of modules to hold subroutines and variables. CPU
acceleration with OpenMP library is in place. Boundary conditions, covering a range of stress, displacement and the hybrid of the two, are supported by DDFS3D....

This repo serves as the official implementation of the paper "DPACS: Hardware Accelerated Dynamic Neural Network Pruning through Algorithm-Architecture Co-design".
DPCAS is an algorithm-architecture co-design framework for dynamic neural network pruning. It utilizes a hardware-aware dynamic spatial and channel pruning mechanism in conjunction with a dynamic dataflow engine in hardware to facilitate efficient processing of the pruned network.
The code is also avaliable on github: https://github.com/CASR-HKU/DPACS
Our group page: https://casr.eee.hku.hk/publication/dpacs-asplos23/

Code and source data for "Echo state graph neural networks with analogue random resistor arrays." The code are tested on Ubuntu 20.04, CUDA 11.1 with PyTorch 1.9.0 and torch-geometric 1.7.2. Both the graph datasets and the source data are included in the `data` folder.

This repository serves as the official code release of the paper "REMOT: A Hardware-Software Architecture for Attention-Guided Multi-Object Tracking with Dynamic Vision Sensors on FPGAs" (FPGA'22).
REMOT is a hardware/software architecture for Multi-Object Tracking using Dynamic Vision Sensors on FPGA. It's designed around the concept of an attention unit (AU). Each AU will only pay attention to a specific region of interest, which is designed to changed as the object moves. In REMOT, a layer...

The self-organizing map (SOM) is a neural network-based classification method with unsupervised learning (Kohonen, 1982). The excellent clustering ability of SOM is valuable because of its noise tolerance and nonlinearity characteristics. This repository provides several Matlab script files that utilize the SOM algorithm for spatiotemporal analysis.The spatiotemporal SOM analysis is applied to explore the spatiotemporal variability in water quality in Hong Kong marine water areas. SOMs are applied at both spatial and temporal domains for...

The self-organizing map (SOM) is a neural network-based classification method with unsupervised learning (Kohonen, 1982). The excellent clustering ability of SOM is valuable because of its noise tolerance and nonlinearity characteristics. This repository provides several Matlab script files that utilize the SOM algorithm for spatiotemporal analysis.The spatiotemporal SOM analysis is applied to explore the spatiotemporal variability in water quality in Hong Kong marine water areas. SOMs are applied at both spatial and temporal domains for...

DDFS3D is a set of open-source code which utilizes DDM (displacement discontinuity method), FSM (fictitious stress method) and the hybrid of DD-FS to efficiently and accurately simulate fractures embedded in large 3D domains. DDFS3D is developed in Fortran language with a number of modules to hold subroutines and variables. CPU
acceleration with OpenMP library is in place. Boundary conditions, covering a range of stress, displacement and the hybrid of the two, are supported by DDFS3D....

Google Earth Engine code for estimating greenspace exposure

Google Earth Engine code for estimating greenspace exposure

Estimation of submarine groundwater discharge (SGD) has traditionally relied on isotopic mass balance calculation, which is a form of regional box modeling. Here we propose an entirely new approach for SGD estimation by coupling a general marine hydrodynamic simulation model with a global optimization algorithm. DYCORS (DYnamic COordinate search using Response Surface models) is a surrogate-based optimization approach for high-dimensional, expensive, and black-box function’s optimization (Regis & Shoemaker, 2013). Because of its effectiveness and adaptability,...

Code and source data for "Echo state graph neural networks with analogue random resistor arrays." The code are tested on Ubuntu 20.04, CUDA 11.1 with PyTorch 1.9.0 and torch-geometric 1.7.2. Both the graph datasets and the source data are included in the `data` folder.