projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 02d1175) | patch
[mlir][sparse][python] migrate more code from boilerplate into proper numpy land
authorAart Bik <ajcbik@google.com>
Thu, 19 Aug 2021 22:50:24 +0000 (15:50 -0700)
committerAart Bik <ajcbik@google.com>
Fri, 20 Aug 2021 16:18:17 +0000 (09:18 -0700)
commit24ea94ad0c1d150a4163577e8f8bd8487edf13ef
treee0f0e3211fd035d451fbe1304d42771c893f839ftree | snapshot
parent02d11757c163feec146cdd21caf9b40cdcdf605fcommit | diff
[mlir][sparse][python] migrate more code from boilerplate into proper numpy land

The boilerplate was setting up some arrays for testing. To fully illustrate
python - MLIR potential, however, this data should also come from numpy land.

Reviewed By: bixia

Differential Revision: https://reviews.llvm.org/D108336
mlir/test/python/dialects/sparse_tensor/test_SpMM.py diff | blob | history
Domain: System / Toolchain;