this project was faster than using minikube containers, and gave me a great throughput of 400k reviews in 3 minutes.
the projects aims to first categorize and then, summarize, these categorized reviews.
the core of this program is the architecture of using dask local cluster to parallelize multiple sentence transformers and summarizers on a single GPU / treating this GPU as a minicluster.
I would like some feedback on this project/architecture, as i have not seen any project do this.
drop by my github and maybe leave a star if you like the project. I am currently working on metrics to compare my architecture with other libaries/ technologies to see which is best for my usecase.