Welcome to SparseBase’s documentation!
Our vision for SparseBase is a library built in C++ that encapsulates, preprocesses, and performs I/O operations on sparse data structures seamlessly and optimally to provide a backbone for algorithms that use these structures.
It is designed with HPC (High Performance Computing) usage at the forefront. It is meant as a container of sparse objects such as tensors, graphs, multi-graphs, and hypergraphs. It mainly focuses on re-ordering, partitioning, and coarsening sparse objects. Also, it supports many different formats (representations) of sparse data and enables smooth conversion between formats.
SparseBase is not meant as a library for carrying out downstream tasks with data it represents. It won’t do graph embedding, tensor-tensor mathematical operations, or factorization, for example. Instead, it is a container and preprocessor that will make carrying out different downstream tasks easier and more standardized. It will give developers quick utilities in terms of converting between formats, preprocessing, and reading/writing.
The library is currently in an early stage of development. As a result the API is not stable and likely to change in the near future.
If you are new to the project, we suggest starting here: pages/getting_started