torch-mlir/include/npcomp
Sean Silva 59a3f46795 Add support for prim.NumToTensor
With this, we can import BERT!
```
pt_util ~/tmp/bert.pt  --import --exported-name=forward \
| npcomp-opt -torch-globalize-object-graph -inline -symbol-dce
```
https://gist.github.com/silvasean/fe7735ff5d065cc9216f7b0346d0e977

The test case here is a bit unconventional -- it isn't actually valid
Python. To figure out how to generate it I had to go search the PyTorch
codebase for "NumToTensor" and work backward. In this case I found
this
[code](649760e5f1/torch/csrc/jit/frontend/ir_emitter.cpp (L464))
which via a wild guess I was able to turn into a test case.

In this case it didn't take me too long, but when doing this kind of
"add a bunch of trivial stuff to bring up a real model", I'm starting to
think that we might skimp on test cases when it's fairly trivial and not
obvious how to test with a small test.
2021-02-26 10:16:56 -08:00
..
Backend/RefJIT Sever C++ level depend on IREE and rebase on exe and python interface. 2020-11-16 21:32:56 -08:00
Conversion Enable building using LLVM_EXTERNAL_PROJECTS. (#152) 2021-01-26 11:43:43 -07:00
Dialect Add support for prim.NumToTensor 2021-02-26 10:16:56 -08:00
Python NFC: Delete npcomp python API and switch to upstream. 2021-01-08 10:46:24 -08:00
RefBackend Bump llvm-project to 444822d77a7fea28aa49edf24533c987efa1b2ee 2020-12-11 14:43:38 -08:00
Typing Bump llvm-project to 444822d77a7fea28aa49edf24533c987efa1b2ee 2020-12-11 14:43:38 -08:00
CMakeLists.txt [RefBackend] Rename "E2E" to RefBackend. 2020-10-07 10:29:48 -07:00
InitAll.h Bump submodule versions. 2020-09-08 13:26:42 -07:00