Skip to content

minor sparse documentation refinement #984

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
cliffburdick merged 2 commits into from
Jun 4, 2025
Merged

minor sparse documentation refinement #984

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8dbe078
minor sparse documentation refinement
aartbik Jun 4, 2025
8f3c771
typo
aartbik Jun 4, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 6 additions & 6 deletions docs_input/basics/sparse_tensor.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,11 +8,11 @@ The implementation is based on the **Universal Sparse Tensor (UST)** type
that uses a tensor format DSL (Domain Specific Language) to describe a vast
space of storage formats. Although the UST type can easily define many common
storage formats (such as dense vectors and matrices, sparse vectors, sparse
matrices in COO, CSR, CSC, DCSR, DCSC, DIA, or BSR format, with generalizations
for sparse tensors), it can also define many less common storage formats.
From MatX's perspective, the advantage of using the UST type (rather than
various specific sparse storage formats) is that the framework code only has
to deal with a single new sparse type (and only dispatch to specific formats
matrices in formats like COO, CSR, CSC, DCSR, DCSC, BSR, BSC, DIA, and with
generalizations to sparse tensors), it can also define many less common storage
formats. From MatX's perspective, the advantage of using the UST type (rather
than various specific sparse storage formats) is that the framework code only
has to deal with a single new sparse type (and only dispatch to specific formats
when required by a high performance library implementation). Also, the tensor
format DSL can be easily extended to include even more sparse storage formats
in the future. From the user's perspective, the UST type provides more
Expand Down Expand Up @@ -248,7 +248,7 @@ block format has 2 dimensions and 4 levels)::

DCSC: (i, j) -> ( j : compressed, i : compressed )

DIA: (i, j) -> ( j - i : compressed, j : range )
DIA-J: (i, j) -> ( j - i : compressed, j : range )

BSR with 2x3 blocks: ( i, j ) -> ( i floordiv 2 : dense,
j floordiv 3 : compressed,
Expand Down