[mlir][sparse] Add complex number reading from files.Support complex numbers for Matrix Market Exchange Formats. Add a test case.Reviewed By: aartbikDifferential Revision: https://reviews.llvm.
[mlir][sparse] Add complex number reading from files.Support complex numbers for Matrix Market Exchange Formats. Add a test case.Reviewed By: aartbikDifferential Revision: https://reviews.llvm.org/D127138
show more ...
[mlir][sparse] remove unused test matrix fileReviewed By: bixiaDifferential Revision: https://reviews.llvm.org/D120069
Accept symmetric sparse matrix in Matrix Market Exchange Format.Reviewed By: aartbikDifferential Revision: https://reviews.llvm.org/D114402
[mlir][sparse] Remove comment w/ code in itReviewed By: aartbikDifferential Revision: https://reviews.llvm.org/D107484
[mlir][sparse] Add Matricized Tensor Times Khatri-Rao Product (MTTKRP) integration testSee this documentation from taco:http://tensor-compiler.org/docs/data_analytics/index.htmlReviewed By: aart
[mlir][sparse] Add Matricized Tensor Times Khatri-Rao Product (MTTKRP) integration testSee this documentation from taco:http://tensor-compiler.org/docs/data_analytics/index.htmlReviewed By: aartbikDifferential Revision: https://reviews.llvm.org/D104417
[mlir][sparse] integration test for all-dense annotated "sparse" outputReviewed By: gussmith23Differential Revision: https://reviews.llvm.org/D104277
[mlir][sparse] support integral types i32,i16,i8 for *numerical* valuesSome sparse matrices operate on integral values (in contrast with the commonf32 and f64 values). This CL expands the compiler
[mlir][sparse] support integral types i32,i16,i8 for *numerical* valuesSome sparse matrices operate on integral values (in contrast with the commonf32 and f64 values). This CL expands the compiler and runtime support to dealwith several common type combinations.Reviewed By: bixiaDifferential Revision: https://reviews.llvm.org/D99999
[mlir][sparse] support for very narrow index and pointer typesRationale:Small indices and values, when allowed by the required range of theinput tensors, can reduce the memory footprint of sparse
[mlir][sparse] support for very narrow index and pointer typesRationale:Small indices and values, when allowed by the required range of theinput tensors, can reduce the memory footprint of sparse tensorseven more. Note, however, that we must be careful zero extendingthe values (since sparse tensors never use negatives for indexing),but LLVM treats the index type as signed in most memory operations(like the scatter and gather). This CL dots all the i's in this regard.Reviewed By: bixiaDifferential Revision: https://reviews.llvm.org/D99777
Move the MLIR integration tests as a subdirectory of test (NFC)This does not change the behavior directly: the tests only run when`-DMLIR_INCLUDE_INTEGRATION_TESTS=ON` is configured. However runni
Move the MLIR integration tests as a subdirectory of test (NFC)This does not change the behavior directly: the tests only run when`-DMLIR_INCLUDE_INTEGRATION_TESTS=ON` is configured. However running`ninja check-mlir` will not run all the tests within a singlelit invocation. The previous behavior would wait for all the integrationtests to complete before starting to run the first regular test. Thetest results were also reported separately. This change is unifying allof this and allow concurrent execution of the integration tests withregular non-regression and unit-tests.Differential Revision: https://reviews.llvm.org/D97241