brusberg/Word2Vec_Kmeans

This code is for the final project for CS-677, Parallel Programming for Many-core Processors. The purpose of this project is to optimize an algorithm/problem with CUDA using AWS EC2 instances. My project tackled clustering Word2Vec word embedding with a k-means implementation. The data I am using for this project is from Stanford's GLOVE project(https://nlp.stanford.edu/projects/glove/), where I am clustering 2.2 million vectors each with 300 dimensions. The overall objective is to see how much time can be reduced from the CPU implementation using a GPU implementation and showing iterations of GPU implementations and how they effect efficiency. I am uploading this before it is due because I would fall apart if my computer died with the code on it and I will update it with the final paper.

Cuda