Projects
A Clang plugin that can be used to turn ordinary C++ functions and objects into command line interfaces (CLIs) similarly to python-fire. It works by adding an extra compiler pass that rewrites source code on the fly in order to inject argument parsing code.
An implemention of the MapReduce programming model written in C++ and based on gRPC. Consists of a single master program that delegates tasks to a number of workers in highly parallel and fault tolerant manner. Based on the MapReduce paper.
Improving mapping of computational tasks to manycore architectures using computational group theory, published in IEEE TCAD. MPsym can determine whether different mappings are equivalent due to symmetries of an architecture’s topology. Works efficiently with large, hierarchical architectures and can be used to improve existing mapping discovery algorithms by making them “symmetry aware”. Implemented in C++ for performance with easy to use Python bindings.
A from-scratch PyTorch implementation of the convolutional neural network described by Zhang et al. in Colorful Image Colorization. Can be used with pretrained weights or trained from scratch. Also includes an implementation of Google’s DeepLab-V3+ network that can be substituted for the default VGG network.
