torch-mlir/lib
Sean Silva 5f3eb637c4 Fix lowering of reduce ops
We were not filling the `outs` with the neutral element of the
reduction, which resulted in reading uninitialized values (we were
getting lucky that sometimes the uninitialized buffers were all zero's).

Also,
- Slight tweak to error messages in the e2e framework.
2021-09-08 15:30:15 -07:00
..
Backend Add TorchToIREE and factor out TorchConversion dialect. 2021-08-16 15:01:58 -07:00
CAPI Import Machine Translation model to MLIR. 2021-08-10 15:22:06 -04:00
Conversion Fix lowering of reduce ops 2021-09-08 15:30:15 -07:00
Dialect Added sum lowering 2021-09-03 17:37:06 -07:00
Interfaces Add AllowsTypeRefinement trait and use it to improve RefineTypes 2021-04-30 10:57:02 -07:00
RefBackend Update llvm-project to 830c0b9023cd0cf91955900e0d96283e7a8c3711 2021-09-03 14:16:38 -07:00
Typing Bump llvm-project to c68d2895a1f4019b387c69d1e5eec31b0eb5e7b0 2021-02-22 12:23:24 -08:00
CMakeLists.txt Add TorchToIREE and factor out TorchConversion dialect. 2021-08-16 15:01:58 -07:00
InitAll.cpp Add TorchToIREE and factor out TorchConversion dialect. 2021-08-16 15:01:58 -07:00