c8d591fb28
[MLIR] import attribute of onnx node ( #383 )
...
* add attributes as NamedAttribute
* support list value for attribute
* use std::tie to avoid c++17 feature
2019-12-21 02:00:58 -05:00
45608282e0
[MLIR] Add support for Relu ( #392 )
...
* Add support for Relu
* Add comments
2019-12-21 01:38:16 -05:00
1c3176bf9f
[MLIR] Lower ONNX element-wise unary ops: Exp, Tanh, Sinh, Cosh, Sigmoid ( #389 )
...
* Lower ExpOp
* Lower ONNXTanhOp
* Lower Exp Tanh, Sinh, and Cosh
* Lower ONNX Sigmoid op
* Merge
* Specialize template lowerScalarOp
* Unify ONNXEWUnaryOpLowering and ONNXEWBinaryOpLowering
* Support multiple types
* Reformat the code
* Add test cases
* Reformat the code
* Change names
* Apply clang-format
* Update variable names
2019-12-21 01:37:29 -05:00
c3ef1d93ae
[MLIR] Lower ONNX element-wise binary ops: Mul, Div, Sub, And, Or, Xor ( #388 )
...
* Lower ONNX element-wise binary ops: Mul, Div, Sub, And, Or, Xor
* Edit gen_doc.py to avoid changes about AnyTypeOf<[AnyMemRef, AnyTensor]>
* Miss a space
* Add tests
* Shorten ONNXElementWiseBinaryOpLowering into ONNXEWBinaryOpLowering
* Move lowering patterns into runOnModule()
* Redundant space
2019-12-21 01:35:31 -05:00
53ab014a1d
[MLIR] Canonicalization pattern for eliminating identity ops ( #377 )
...
* Canonicalization pattern for eliminating identity ops
* Add a test for the identity elimination rule
* Remove frontend from test
* Use CHECK-NEXT instead of CHECK
2019-12-21 00:47:22 -05:00
3f68c5420d
[MLIR] generate op from onnx document ( #366 )
...
* generate op from onnx document
* Restore FullGemm
* update the op attribute for shape inference and canonicalizer
* Update onnx_canonicalization.mlir
2019-12-21 00:40:40 -05:00
TONG CHEN
TUNG LEDUC