qlty: handling large tensors in scientific imaging
This addresses memory constraints for researchers in scientific imaging using deep learning, but it is incremental as it focuses on tooling rather than fundamental advances.
The paper tackles the problem of handling large volumetric datasets in scientific imaging that exceed GPU memory capacity by introducing qlty, a toolkit for tensor management, which enables effective training and inference in resource-limited environments.
In scientific imaging, deep learning has become a pivotal tool for image analytics. However, handling large volumetric datasets, which often exceed the memory capacity of standard GPUs, require special attention when subjected to deep learning efforts. This paper introduces qlty, a toolkit designed to address these challenges through tensor management techniques. qlty offers robust methods for subsampling, cleaning, and stitching of large-scale spatial data, enabling effective training and inference even in resource-limited environments.