Commit Graph

Select branches

Hide Pull Requests

master

Mono Color

• 1622b9f161
  [NFC] Change ONNF based names to ONNX-MLIR (#32) Gheorghe-Teodor Bercea 2020-03-17 09:16:33 -0400
• c25831094e Revert "Support attribute promotion." Tian Jin 2020-03-17 17:41:59 +0800
• 955968b750 Support attribute promotion. Tian Jin 2020-03-17 17:39:34 +0800
• d86591d61a
  Import all initialized tensors as dense constants (#30) Tung D. Le 2020-03-17 00:17:28 +0900
• c46880d5c6
  Fix reshape output shape inference when a single dynamic shape is given (#22) Gheorghe-Teodor Bercea 2020-03-13 17:18:46 -0400
• 6137fc7c17
  Fix issues #15 and #16 (#29) chentong319 2020-03-13 10:19:27 -0400
• 362491553c
  Shape inference for ONNXAveragePool (#21) Tung D. Le 2020-03-13 22:59:16 +0900
• a65820940c
  Lower ConstantOp (#28) Tung D. Le 2020-03-12 23:58:42 +0900
• 162ac1bc32
  Pad value for MaxPool must be negative infinity instead of zero (#20) Tung D. Le 2020-03-12 22:30:02 +0900
• 811b63e031
  Inter common pad (#26) Alexandre Eichenberger 2020-03-11 18:36:02 -0400
• 391f565a66
  Lower constant padding operation to KRNL dialect (#27) chentong319 2020-03-11 16:54:07 -0400
• e8a0b47e10
  Fix case for upper and lower padding when strides are present. (#11) Gheorghe-Teodor Bercea 2020-03-10 16:58:05 -0400
• fe3279e721
  Initialize operation arguments with ONNX model constants (#8) Gheorghe-Teodor Bercea 2020-03-10 14:46:35 -0400
• ba02b90e0b
  Enable inference for arbitrary number of instructions (#12) Gheorghe-Teodor Bercea 2020-03-10 14:16:03 -0400
• 1882059ac9
  Support Pads for MaxPoolSingleOut (#14) Tung D. Le 2020-03-10 09:15:58 +0900
• 718ec85479
  Change variant repo from git to https. (#17) Tian Jin 2020-03-10 00:16:43 +0800
• 8a992b619f
  Create some helper functions to emit constant op for a specific type (#7) Gheorghe-Teodor Bercea 2020-03-05 14:21:00 -0500
• 8e1b30e133
  Check channel dimension mismatch only for known dimensions (#2) Gheorghe-Teodor Bercea 2020-03-04 14:34:08 -0500
• e4c23da4fd
  Lower MaxPoolSingleOutOp to Krnl dialect (#1) Gheorghe-Teodor Bercea 2020-03-04 14:27:21 -0500
• e97df0b343
  Add a pass to decompose ONNX operations (#9) Tung D. Le 2020-03-05 00:53:59 +0900
• 7c1dd0279b
  Merge pull request #5 from tjingrant/update-buildbot Gheorghe-Teodor Bercea 2020-03-02 11:59:54 -0500
• 47831749ce Use the exact commit id specified in clang-ykt/llvm-project. Tian Jin 2020-03-03 00:18:59 +0800
• 04dd904ca7 Switch to new status badge. Tian Jin 2020-03-02 20:37:33 +0800
• 473fdd726d Fix DocCheck error. Tian Jin 2020-03-02 17:06:40 +0800
• 2f5d65f2e4 Invalidate cache. Tian Jin 2020-03-02 16:24:15 +0800
• ee96ffab73 Merge branch 'update-buildbot' of https://github.com/tjingrant/onnx-mlir into update-buildbot Tian Jin 2020-03-02 16:21:10 +0800
• d8b5e195d2 Upgrade MLIR commit id. Tian Jin 2020-03-02 16:20:58 +0800
• 5e2a02ecdf
  Trigger buildbot Tian Jin 2020-03-02 15:00:31 +0800
• f856f84c55 Use llvm-project we know that works. Tian Jin 2020-03-02 14:28:36 +0800
• 5357fc1421
  Use SqrtOp in Standard dialect (#108) Tung D. Le 2020-02-27 02:03:24 +0900
• 0c4a010283
  Remove rank constraints in gemm fusion (#101) Tung D. Le 2020-02-27 01:40:52 +0900
• 24d89625e3
  Remove redundant lower_frontend_to_krnl since we reorganized it (#99) Tung D. Le 2020-02-26 17:32:06 +0900
• 04d9e2f341
  Merge pull request #84 from chentong319/shapeinference-pad chentong319 2020-02-25 19:32:34 -0500
• 4edc97f3de Merge branch 'shapeinference-pad' of github.com:chentong319/ONNF into shapeinference-pad chentong 2020-02-25 17:46:44 -0500
• 3abbf1c0e9 put the common code into a helper function chentong 2020-02-25 17:43:49 -0500
• 4079ee1f26 Merge remote-tracking branch 'upstream/master' into shapeinference-pad chentong 2020-02-25 15:54:18 -0500
• 3a88361b17
  use input/output operation names, use helper for attribute function and int values (#106) Alexandre Eichenberger 2020-02-25 15:46:11 -0500
• 3b1c29c078
  Using attribute setters for maxpool (#105) Alexandre Eichenberger 2020-02-25 14:33:48 -0500
• e02aa87748
  Update gitignore file to ignore Filesystem artifacts and python related temporary files. (#103) Tian Jin 2020-02-26 00:18:37 +0800
• 907104d7e8
  Merge branch 'master' into shapeinference-pad Gheorghe-Teodor Bercea 2020-02-25 11:14:28 -0500
• ee3e140ddb
  [NFC] Change structure of conversion folder. (#96) Gheorghe-Teodor Bercea 2020-02-25 10:38:08 -0500
• 32f08bcf0c
  Clean-up code. (#98) Gheorghe-Teodor Bercea 2020-02-25 09:54:29 -0500
• 0d307d1183
  Set flag to true when definition is emitted. (#97) Gheorghe-Teodor Bercea 2020-02-25 09:47:42 -0500
• a720f9a7b2
  Remove special GemmNoBias since we can handle it using NoneType bias (#100) Tung D. Le 2020-02-25 14:20:43 +0900
• 732317cd5a
  Transition to ONNX-1.6.0. (#95) Tian Jin 2020-02-25 13:04:15 +0800
• 1ad7989fc5
  Merge branch 'master' into shapeinference-pad Gheorghe-Teodor Bercea 2020-02-24 17:22:00 -0500
• fcb5f35993
  Introduce helper class to generate KRNL code and apply it to Convolution (#93) Alexandre Eichenberger 2020-02-24 17:20:15 -0500
• d4f8fef947
  Merge branch 'master' into shapeinference-pad Gheorghe-Teodor Bercea 2020-02-24 16:13:21 -0500
• 9c398c0121
  Support Optional Inputs (#94) Tian Jin 2020-02-24 23:46:48 +0800
• b3df3c64b5 Merge branch 'master' of github.com:clang-ykt/ONNF into shapeinference-pad chentong 2020-02-24 09:26:45 -0500
• 5ab7fe37c4
  Merge branch 'master' into shapeinference-pad chentong319 2020-02-21 09:36:41 -0500
• 2281cc060f Merge branch 'master' of github.com:clang-ykt/ONNF into shapeinference-pad chentong 2020-02-21 09:30:40 -0500
• 479dd5e35a
  Add result type inference to op definition (#87) Tung D. Le 2020-02-21 23:28:24 +0900
• aea6479ad3
  Lower BatchNormalization (test mode) to Krnl dialect (#70) Tung D. Le 2020-02-21 01:45:40 +0900
• f1d20e368f
  Add support of GemmNoBias (#91) Tung D. Le 2020-02-21 00:55:24 +0900
• a3f042220e
  Get MemRefType for result types (#69) Tung D. Le 2020-02-20 22:44:02 +0900
• b28c6906b4
  Fix building ONNF with latest LLVM/MLIR (#89) Gheorghe-Teodor Bercea 2020-02-19 18:15:02 -0500
• da037ffc7d
  Merge branch 'master' into shapeinference-pad Gheorghe-Teodor Bercea 2020-02-19 13:46:00 -0500
• b9f2f25b56
  [NFC] Categorize ONNX ops lowering (#80) Tung D. Le 2020-02-19 16:17:48 +0900
• c11f97f1b5 fix test case for changes in Op definition chentong 2020-02-17 09:07:58 -0500
• ec43fadc3b Merge remote-tracking branch 'upstream/master' into shapeinference-pad chentong 2020-02-17 08:27:43 -0500
• 3c505ae31d
  Split convolution into explicit padding and unpaded convolution. (#82) Gheorghe-Teodor Bercea 2020-02-14 16:06:38 -0500
• bbdf4e3b4d Merge remote-tracking branch 'upstream/master' into shapeinference-pad chentong 2020-02-14 15:35:47 -0500
• 17d84901b7
  Allow 1-D convolutions. (#86) Gheorghe-Teodor Bercea 2020-02-14 10:54:08 -0500
• b521719587
  Lower Matmul operation to Krnl dialect (#57) Tung D. Le 2020-02-15 00:43:17 +0900
• 960de50b7d Merge branch 'master' of github.com:clang-ykt/ONNF into shapeinference-pad chentong 2020-02-13 20:00:17 -0500
• c3041bfb43 shape inference for pad with constant pads chentong 2020-02-13 12:08:29 -0500
• dab862e4f1
  Update README.md Gheorghe-Teodor Bercea 2020-02-13 15:52:53 -0500
• ecf16c3441
  Update README.md Gheorghe-Teodor Bercea 2020-02-13 15:51:39 -0500
• 8b9fb2f49f
  Update README.md Gheorghe-Teodor Bercea 2020-02-13 13:46:39 -0500
• 7d5c33448e
  Update README.md Gheorghe-Teodor Bercea 2020-02-13 13:43:52 -0500
• 5587870da4
  Update README.md Gheorghe-Teodor Bercea 2020-02-13 13:43:19 -0500
• 937bbec265
  Fix scalar entry point parameter lowering issue. (#78) Tian Jin 2020-02-13 13:50:05 +0800
• e5677bba1f
  Add e2e tests related to Reduction Ops (#79) Tung D. Le 2020-02-12 17:40:32 +0900
• 49dae74eab
  Create constant pad (#75) chentong319 2020-02-11 15:32:01 -0500
• 094be4f37a
  Add support for strides when emitting convolution loop nest. (#76) Gheorghe-Teodor Bercea 2020-02-11 11:53:13 -0500
• adad9e24bd
  Add support of negative dimensions (#66) Tung D. Le 2020-02-12 00:37:47 +0900
• 181803ebf4
  Using Tablegen to Generate Op Documentation (#74) Tian Jin 2020-02-11 03:18:54 +0800
• 2c7046ff5f
  Lowering ReductionMax, ReductionMin, ReductionProd and ReductionSum (#31) Tung D. Le 2020-02-10 22:38:19 +0900
• 0272451521
  Lower convolution to KRNL dialect. (#65) Gheorghe-Teodor Bercea 2020-02-07 16:51:32 -0500
• 0564c0eaef
  Add constraints for matmul-add fusion (#67) Tung D. Le 2020-02-08 03:51:44 +0900
• 0bfb660d02
  Import 2-argument Gemm as GemmNoBias (#68) Tung D. Le 2020-02-08 03:45:37 +0900
• 60ac8f081f
  Op def output (#73) chentong319 2020-02-07 11:10:35 -0500
• ae297f14ee
  Revert "fix name of operator output in onnxop.inc and Operator.md (#62)" (#72) Gheorghe-Teodor Bercea 2020-02-06 10:52:57 -0500
• c45655413d
  fix name of operator output in onnxop.inc and Operator.md (#62) chentong319 2020-02-06 07:54:03 -0500
• 477227a0ec
  Added lowering of SignOp (#21) Haruki Imai 2020-02-04 23:27:17 +0900
• 87aa72764f
  Fix dependencies for onnf-opt (#51) Gheorghe-Teodor Bercea 2020-01-31 10:51:43 -0500
• 5791cfdce7
  Resolve buildbot issue. (#64) Tian Jin 2020-01-31 23:24:45 +0800
• 2b56c09454
  Rewrite ReduceL1, ReduceL2, ReduceLogSum, ReduceLogSumExp, ReduceSumSquare in the ONNX dialect (#38) Tung D. Le 2020-01-31 20:00:39 +0900
• 0d77840969
  Inference maxpool (#48) Alexandre Eichenberger 2020-01-30 14:30:28 -0500
• 9fb826ae7e
  Lower transpose operation to KRNL dialect (#54) Gheorghe-Teodor Bercea 2020-01-30 11:44:56 -0500
• 6959cf4586
  clean up gen_doc.py (#59) chentong319 2020-01-29 13:54:46 -0500
• ce8594fc60
  Create ImportONNXDefs.md (#58) chentong319 2020-01-29 13:45:48 -0500
• 400676e371
  Lowering Gemm (#19) Tung D. Le 2020-01-30 01:11:49 +0900
• 9e82d388f0
  Add support for Unsqueeze (#50) Tung D. Le 2020-01-30 00:46:02 +0900
• 5b44169aaa
  Support dimension zero in reshape (#55) Tung D. Le 2020-01-30 00:41:09 +0900
• f3047943a1
  Handle 1-D MATMUL N-D (#56) Tung D. Le 2020-01-30 00:35:05 +0900
• 195bf9d15d Add KrnlSqrtOp (#22) Tung D. Le 2020-01-29 01:10:47 +0900
• f00206cecf
  Fix reshape op. (#53) Gheorghe-Teodor Bercea 2020-01-28 10:21:08 -0500
• 7c889548a7 Allow importing variadic inputs/outputs of onnx operators (#16) Tung D. Le 2020-01-28 22:48:11 +0900