colin
/
onnx-mlir
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
onnx-mlir/src/Builder/OpBuildTable.inc

319 lines
11 KiB
PHP
Raw Normal View History

clean up gen_doc.py (#59) Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-01-30 02:54:46 +08:00
//********************************************************
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
// Do not modify this file directly.
// This file is automatically generated via script.
Setup Documentation Page using Github Pages (#76) * Setup documentation server, move doc files from /doc to /docs as per Github Pages convention. * Include deleted files in patch. * /doc -> /docs * /doc -> /docs * Update documentation on importing ONNX spec into ONNX Dialect; provide documentation on how to add new documentation pages.
2020-04-09 23:37:04 +08:00
// Details can be found in docs/readonnxdefs.md .
clean up gen_doc.py (#59) Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-01-30 02:54:46 +08:00
//********************************************************
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Abs")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXAbsOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Acos")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXAcosOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Acosh")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXAcoshOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Add")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXAddOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "And")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXAndOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ArgMax")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXArgMaxOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ArgMin")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXArgMinOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Asin")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXAsinOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Asinh")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXAsinhOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Atan")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXAtanOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Atanh")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXAtanhOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "AveragePool")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXAveragePoolOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "BatchNormalization")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
ImportNodeBatchNormalization(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "BitShift")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXBitShiftOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Cast")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXCastOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Ceil")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXCeilOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Clip")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXClipOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Compress")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXCompressOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Concat")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXConcatOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ConcatFromSequence")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXConcatFromSequenceOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Constant")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXConstantOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ConstantOfShape")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXConstantOfShapeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Conv")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXConvOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ConvInteger")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXConvIntegerOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ConvTranspose")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXConvTransposeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Cos")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXCosOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Cosh")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXCoshOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "CumSum")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXCumSumOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "DepthToSpace")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXDepthToSpaceOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "DequantizeLinear")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXDequantizeLinearOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Det")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXDetOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Div")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXDivOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Dropout")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXDropoutOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "DynamicQuantizeLinear")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXDynamicQuantizeLinearOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Elu")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXEluOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Equal")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXEqualOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Erf")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXErfOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Exp")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXExpOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Expand")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXExpandOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "EyeLike")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXEyeLikeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Flatten")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXFlattenOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Floor")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXFloorOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "GRU")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXGRUOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Gather")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXGatherOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "GatherElements")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXGatherElementsOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "GatherND")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXGatherNDOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Gemm")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXGemmOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "GlobalAveragePool")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXGlobalAveragePoolOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "GlobalLpPool")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXGlobalLpPoolOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "GlobalMaxPool")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXGlobalMaxPoolOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Greater")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXGreaterOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "HardSigmoid")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXHardSigmoidOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Hardmax")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXHardmaxOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Identity")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXIdentityOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "If")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXIfOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "InstanceNormalization")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXInstanceNormalizationOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "IsInf")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXIsInfOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "IsNaN")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXIsNaNOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "LRN")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXLRNOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "LSTM")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXLSTMOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "LeakyRelu")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXLeakyReluOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Less")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXLessOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Log")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXLogOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "LogSoftmax")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXLogSoftmaxOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Loop")
Implement PadOp based on attribute promotion (#71) * enable promote attr for pad * use optional arguments for pad * shape infereance for pad * Lowering Pad * format file * use DenseTensor for the attribute * use Pad in ONNXRewrite * fix the merge conflict * fix the attr given to constantOp * handle ONNXConstantOp in attribute promotion * Fix bug when AttributePromotion is called more than once * update ONNXOps.td.inc with correct version of onnx * update onnx.md * responses to review * fix the build error * change the implementation of Pad * delete commented out code * clang format Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-05-15 13:19:28 +08:00
buildOperation<mlir::ONNXLoopOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "LpNormalization")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXLpNormalizationOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "LpPool")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXLpPoolOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "MatMul")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXMatMulOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "MatMulInteger")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXMatMulIntegerOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Max")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXMaxOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "MaxPool")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
ImportNodeMaxPool(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "MaxRoiPool")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXMaxRoiPoolOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "MaxUnpool")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXMaxUnpoolOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Mean")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXMeanOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "MeanVarianceNormalization")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXMeanVarianceNormalizationOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Min")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXMinOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Mod")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXModOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Mul")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXMulOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Multinomial")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXMultinomialOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Neg")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXNegOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "NonMaxSuppression")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXNonMaxSuppressionOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "NonZero")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXNonZeroOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Not")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXNotOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "OneHot")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXOneHotOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Or")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXOrOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "PRelu")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXPReluOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Pad")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
ImportNodePad(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Pow")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXPowOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "QLinearConv")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXQLinearConvOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "QLinearMatMul")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXQLinearMatMulOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "QuantizeLinear")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXQuantizeLinearOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "RNN")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXRNNOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "RandomNormal")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXRandomNormalOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "RandomNormalLike")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXRandomNormalLikeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "RandomUniform")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXRandomUniformOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "RandomUniformLike")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXRandomUniformLikeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Range")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXRangeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Reciprocal")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReciprocalOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReduceL1")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReduceL1Op>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReduceL2")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReduceL2Op>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReduceLogSum")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReduceLogSumOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReduceLogSumExp")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReduceLogSumExpOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReduceMax")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReduceMaxOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReduceMean")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReduceMeanOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReduceMin")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReduceMinOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReduceProd")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReduceProdOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReduceSum")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReduceSumOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReduceSumSquare")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReduceSumSquareOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Relu")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReluOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Reshape")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
ImportNodeReshape(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Resize")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXResizeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ReverseSequence")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXReverseSequenceOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "RoiAlign")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXRoiAlignOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Round")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXRoundOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Scan")
Implement PadOp based on attribute promotion (#71) * enable promote attr for pad * use optional arguments for pad * shape infereance for pad * Lowering Pad * format file * use DenseTensor for the attribute * use Pad in ONNXRewrite * fix the merge conflict * fix the attr given to constantOp * handle ONNXConstantOp in attribute promotion * Fix bug when AttributePromotion is called more than once * update ONNXOps.td.inc with correct version of onnx * update onnx.md * responses to review * fix the build error * change the implementation of Pad * delete commented out code * clang format Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-05-15 13:19:28 +08:00
buildOperation<mlir::ONNXScanOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Scatter")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXScatterOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ScatterElements")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXScatterElementsOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ScatterND")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXScatterNDOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Selu")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSeluOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "SequenceAt")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSequenceAtOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "SequenceConstruct")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSequenceConstructOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "SequenceEmpty")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSequenceEmptyOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "SequenceErase")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSequenceEraseOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "SequenceInsert")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSequenceInsertOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "SequenceLength")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSequenceLengthOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Shape")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXShapeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Shrink")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXShrinkOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Sigmoid")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSigmoidOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Sign")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSignOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Sin")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSinOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Sinh")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSinhOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Size")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSizeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Slice")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSliceOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Softmax")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSoftmaxOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Softplus")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSoftplusOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Softsign")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSoftsignOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "SpaceToDepth")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSpaceToDepthOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Split")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSplitOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "SplitToSequence")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSplitToSequenceOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Sqrt")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSqrtOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Squeeze")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSqueezeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "StringNormalizer")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXStringNormalizerOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Sub")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSubOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Sum")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXSumOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Tan")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXTanOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Tanh")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXTanhOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "TfIdfVectorizer")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXTfIdfVectorizerOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "ThresholdedRelu")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXThresholdedReluOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Tile")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXTileOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "TopK")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXTopKOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Transpose")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXTransposeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Unique")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXUniqueOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Unsqueeze")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXUnsqueezeOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Upsample")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXUpsampleOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Where")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXWhereOp>(node);
Support Optional Inputs (#94) * 1. Combine variadicIn/Out with expectedNumOperands/Results to simplify import function arguments. 2. Generic improvements to code readability in gen_doc.py. * Update ONNX Dialect doc. * Remove redundant code in ImportNode. * Prettify op_build_table.inc. * 1. Remove irrelevant code in gen_doc.py * Refactor code to be more readable. * Further refactoring for readability improvements. * Allow gemm to have an optional operand (bias term), and include an example of declarative optimization pattern targeting gemm with bias term ommitted. * Make shape inference/lowering of gemm op compatible with optional operand declaration. * Apply canonicalization again after lowering from onnx -> std dialects. * Make hasBias compatible with the situation of GemmNoBias op. * Update doc. * Add a canonicalization test. * Remove special handler for importing Gemm op, as it's redundant now.
2020-02-24 23:46:48 +08:00
if (opName == "Xor")
Express some basic features of an Operation in TableGen file (#103) * change operation definition * change importer * default type inference * file format * generate types for input/output * generate the mapping for operation output type * remove debug message for gen_doc.py * update the dialect doc * add support Complex * format * update document Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-05-22 10:03:16 +08:00
buildOperation<mlir::ONNXXorOp>(node);