platform/upstream/tvm.git
5 years ago[TOPI] Use cblas for dense and batch_matmul when "cblas" is in the target libraries...
Jon Soifer [Wed, 21 Aug 2019 05:24:10 +0000 (22:24 -0700)]
[TOPI] Use cblas for dense and batch_matmul when "cblas" is in the target libraries (#3787)

* Support cblas library in dense

* start to add support for generic batch_matmul compute

* Add x86 override for batch_matmul

* Fix linting

* reset file

* Fix typos

* dummy change to re-trigger CI

5 years ago[Relay][VM]VM Profiler (#3727)
Wei Chen [Wed, 21 Aug 2019 04:28:05 +0000 (21:28 -0700)]
[Relay][VM]VM Profiler (#3727)

* [Relay][VM]VM debugger

* Report mean/min/max for op duration

* Typos

* Lint

* Lint

* Lint

* Support build debug VM in CMake

* Lint

* Enable VM debug in unit test

* Disable debug vm test until new docker image is built

* Add device sync code

* Fix qnn unit test

* Disable vm debug by default

* Rename files

* Rename classes

* Fix comment

* Fix comment

5 years agoadd gfx906 bc (#3808)
Peter Yeh [Wed, 21 Aug 2019 01:57:01 +0000 (18:57 -0700)]
add gfx906 bc (#3808)

5 years ago[dep] psutil (#3780)
Bing Xu [Tue, 20 Aug 2019 21:45:18 +0000 (14:45 -0700)]
[dep] psutil (#3780)

5 years ago[CI] Temporary disable rust test (#3809)
Tianqi Chen [Tue, 20 Aug 2019 21:44:23 +0000 (05:44 +0800)]
[CI] Temporary disable rust test (#3809)

5 years ago[CI] Solve occasional CI issue when pad value is all 0 (#3801)
Zhao Wu [Mon, 19 Aug 2019 21:50:49 +0000 (05:50 +0800)]
[CI] Solve occasional CI issue when pad value is all 0 (#3801)

5 years ago[Relay][Frontend][TFLite] transpose implementation for tflite.py (#3705)
Neo Chien [Mon, 19 Aug 2019 17:20:38 +0000 (01:20 +0800)]
[Relay][Frontend][TFLite] transpose implementation for tflite.py (#3705)

* transpose implementation for tflite.py

* add TRANSPOSE to convert_map

* Fix Unexpected keyword argument 'axis' in function call

* add test for transpose oprator

* Add the parameter 'axes' handling

* add test for transpose oprator

* solve conflict within CONTRIBUTORS.md

* Improve the if condition for empty tuple

* Add one unit test to cover empty tuple

* solve conflict within CONTRIBUTORS.md

5 years agoFixed onnx test failures when run on a cpu backend (#3764)
tristan-arm [Mon, 19 Aug 2019 16:33:43 +0000 (17:33 +0100)]
Fixed onnx test failures when run on a cpu backend (#3764)

* Fixed onnx test failures when run on a cpu backend

* Updated check_torch_conversion function to include output comparison

5 years ago[CoreML] Solve CoreML frontend issue of image scaler and padding so that Mobilenet...
Zhao Wu [Mon, 19 Aug 2019 06:05:41 +0000 (14:05 +0800)]
[CoreML] Solve CoreML frontend issue of image scaler and padding so that Mobilenet mlmodel can work correctly. (#3800)

5 years ago[TOPI, CUDA] Improve conv2d_transpose schedule template (#3796)
Wuwei Lin [Mon, 19 Aug 2019 05:58:09 +0000 (01:58 -0400)]
[TOPI, CUDA] Improve conv2d_transpose schedule template (#3796)

5 years ago[VTA][TSIM] parallel TSIM hardware compilation with macOS and debug support (#3797)
Liangfu Chen [Sun, 18 Aug 2019 23:19:22 +0000 (07:19 +0800)]
[VTA][TSIM] parallel TSIM hardware compilation with macOS and debug support (#3797)

* [VTA][TSIM] parallel hardware compilation with macOS and debug support

* simplify

5 years ago[Community] Hao Lu -> Committer (#3789)
ziheng [Sat, 17 Aug 2019 07:51:42 +0000 (00:51 -0700)]
[Community] Hao Lu -> Committer (#3789)

5 years agoFix ArgBinder assert order (#3794)
Wuwei Lin [Sat, 17 Aug 2019 03:28:46 +0000 (20:28 -0700)]
Fix ArgBinder assert order (#3794)

5 years ago[BUGFIX] Fix for NoneType Target (#3792)
ziheng [Fri, 16 Aug 2019 20:14:29 +0000 (13:14 -0700)]
[BUGFIX] Fix for NoneType Target (#3792)

5 years ago[Relay][Quantization] Fix out-of-date realize (#3790)
Wuwei Lin [Fri, 16 Aug 2019 06:23:08 +0000 (23:23 -0700)]
[Relay][Quantization] Fix out-of-date realize (#3790)

5 years agoQNN quantize and dequantize operators. (#3745)
shoubhik [Fri, 16 Aug 2019 03:29:34 +0000 (20:29 -0700)]
QNN quantize and dequantize operators. (#3745)

* QNN quantize and dequantize operators.

* addressing review comments.

* addressing review comments.

* Adding new line at the end of the file.

* Adhering to styling guidelines.

* Adding name to contributors.

* Fixing lint issue.

* Fixing file name.

* Removing unnecessary code.

5 years ago[Relay][Frontend][ONNX] Add Sign and Equal operators to ONNX frontend (#3760)
Jon Soifer [Thu, 15 Aug 2019 18:41:54 +0000 (11:41 -0700)]
[Relay][Frontend][ONNX] Add Sign and Equal operators to ONNX frontend (#3760)

* [Relay][Frontend][ONNX] Add Sign and Equal operators to ONNX frontend

* Dummy change to retrigger integration test

5 years ago[QUANTIZE] Refactor quantization codebase and fix model accuracy (#3543)
ziheng [Thu, 15 Aug 2019 09:31:30 +0000 (02:31 -0700)]
[QUANTIZE] Refactor quantization codebase and fix model accuracy (#3543)

* Refactor.

* update

* update

* update

* update

* update

* update

5 years ago[QNN] InferType changes that missed CI. (#3779)
Animesh Jain [Thu, 15 Aug 2019 06:35:15 +0000 (23:35 -0700)]
[QNN] InferType changes that missed CI. (#3779)

5 years ago[QNN] Concatenate operator (#3730)
Animesh Jain [Wed, 14 Aug 2019 23:56:58 +0000 (16:56 -0700)]
[QNN] Concatenate operator (#3730)

5 years ago[Relay][Legalize][ARM_CPU] Handling NHWC layout for arm_cpu. (#3754)
Animesh Jain [Wed, 14 Aug 2019 23:44:13 +0000 (16:44 -0700)]
[Relay][Legalize][ARM_CPU] Handling NHWC layout for arm_cpu. (#3754)

5 years ago[VTA][Chisel] scale dram base address in hardware instead of runtime (#3772)
Luis Vega [Wed, 14 Aug 2019 17:42:59 +0000 (10:42 -0700)]
[VTA][Chisel] scale dram base address in hardware instead of runtime (#3772)

* [VTA][Chisel] scale dram base address in hardware instead of runtime

* remove trailing spaces

5 years agofix dense tuning (#3768)
Thierry Moreau [Wed, 14 Aug 2019 07:17:19 +0000 (00:17 -0700)]
fix dense tuning (#3768)

5 years agofix (#3769)
雾雨魔理沙 [Wed, 14 Aug 2019 05:31:18 +0000 (22:31 -0700)]
fix (#3769)

5 years ago[Relay][Frontend][TensorFlow] Support BatchMatMul with input dimensions larger than...
Jon Soifer [Wed, 14 Aug 2019 00:15:45 +0000 (17:15 -0700)]
[Relay][Frontend][TensorFlow] Support BatchMatMul with input dimensions larger than 3 (#3732)

* Support BatchMatMul with shapes greater than length 3

* Fixes

* Add tests

* Remove dependency on Python3

* Clean up

* Merge with master

* Resolve comments

5 years agosyntax fix (#3765)
Benjamin Tu [Tue, 13 Aug 2019 23:03:13 +0000 (16:03 -0700)]
syntax fix (#3765)

5 years ago[VTA][Chisel] run all unittests by default (#3766)
Luis Vega [Tue, 13 Aug 2019 23:02:48 +0000 (16:02 -0700)]
[VTA][Chisel] run all unittests by default (#3766)

* [VTA][Chisel] run all unittests by default

* better naming

* add generated unittest folder to clean rule

5 years agofix some pass docs (#3767)
Zhi [Tue, 13 Aug 2019 22:23:50 +0000 (15:23 -0700)]
fix some pass docs (#3767)

5 years ago[VTA][TSIM][Build] Towards TSIM CI testing (#3704)
Thierry Moreau [Tue, 13 Aug 2019 21:01:46 +0000 (14:01 -0700)]
[VTA][TSIM][Build] Towards TSIM CI testing (#3704)

* building TSIM specific library along with fast simulator to quickly switch between dlls

* cmake controlled TSIM libraries

* always build tsim driver in either simulation modes

* build DLLs based on CMAKE flags

* updating the jenkinsfile

* small restructuring

* reducing the cmake flags

* update instructions

* reverting to 3 flags

* update Jenkinsfile

* adding new line

* enabling TSIM unit and integration tests

* fix description

* temporarily disabling task_python_vta tests in CPU Build stage

* move CPU tests in unit test stage

* stage  reorg

* better make

* disabling TSIM tests for now

* reverting some restructuring

* fix

5 years ago[VTA] [Chisel] Improved Data Gen, Added ALU Test (#3743)
Benjamin Tu [Tue, 13 Aug 2019 16:58:32 +0000 (09:58 -0700)]
[VTA] [Chisel] Improved Data Gen, Added ALU Test (#3743)

* added alutest

* fix indent

* name change for cycle

* improved data gen and infra

* added alutest

* fix indent

* name change for cycle

* improved data gen and infra

* fix space

* fix indent

* fixes

* aluRef

* fix randomarary

* add

* Revert "add"

This reverts commit 87077daebbe055dee11f80e37da3a6291138e0f0.

* Revert "fix randomarary"

This reverts commit df386c1e660eb6ebcff1a1f905610573676f1589.

* Revert "aluRef"

This reverts commit 8665f0d4a7b12b796b2cb1ca6bf9cfe5613ee389.

* should fix dlmc-core

5 years agofix mistype (#3763)
pzq [Tue, 13 Aug 2019 16:37:50 +0000 (00:37 +0800)]
fix mistype (#3763)

5 years ago [ARITH] Simplify casts of constants 0 and 1 (#3758)
Sergei Grechanik [Tue, 13 Aug 2019 15:28:28 +0000 (18:28 +0300)]
 [ARITH] Simplify casts of constants 0 and 1 (#3758)

* [ARITH] Simplify casts of constants 0 and 1

* [EXPR] is_const_value to check whether non-ints are consts

* Revert "[EXPR] is_const_value to check whether non-ints are consts"

This reverts commit 7e1b3462e3f74fd0afb1541d72978107cfa23c30.

* Use tvm::cast

5 years ago[Relay] SpaceToDepth and MirrorPad Operators (#3718)
Josh Fromm [Tue, 13 Aug 2019 00:48:10 +0000 (17:48 -0700)]
[Relay] SpaceToDepth and MirrorPad Operators (#3718)

* Added relay and topi mirror_pad operator.

* Added mirror_padding to tensorflow frontend.

* Added mirrorpad testing in tensorflow frontent.

* Added space_to_depth in tf frontend.

* Added tests for spacetodepth.

* spacetodepth bug fix.

* Lint fix

* Added mirror pad python attrs.

* Pad code formatting.

* Syntax improvement

* Hopefully last lint fix

5 years agoDon't replace reduction init axis with new axis if bound to a thread. (#3408)
Christian Sarofeen [Mon, 12 Aug 2019 21:11:11 +0000 (17:11 -0400)]
Don't replace reduction init axis with new axis if bound to a thread. (#3408)

* Don't replace reduction init axis with new axis if bound to a thread.

* Linter.

* Reduce bind test case.

* Guard test on CUDA support.

* [CUDA TE TESTS] Add rfactor predicate test, add global bx and tx.

* [CUDA TE TESTS] Add loop partition test for simple rfactor case.

5 years ago[Bugfix] tvm.scan follow by tvm.compute segfault (#3723)
Yizhi Liu [Mon, 12 Aug 2019 16:04:48 +0000 (00:04 +0800)]
[Bugfix] tvm.scan follow by tvm.compute segfault (#3723)

* [bugfix] tvm.scan follow by tvm.compute segfault

* more strict bound condition check

* access k + 1 -> k

* fix scan test

5 years agoFix the potential index overflow (#3751)
Neo Chien [Mon, 12 Aug 2019 15:54:18 +0000 (23:54 +0800)]
Fix the potential index overflow (#3751)

5 years agoRevert compile_cmd kwarg name change (#3746)
Logan Weber [Mon, 12 Aug 2019 15:52:19 +0000 (08:52 -0700)]
Revert compile_cmd kwarg name change (#3746)

* Revert compile_cmd kwarg name change

* Fix binutil tests

5 years agoadd reviewer (#3755)
Haichen Shen [Mon, 12 Aug 2019 15:52:10 +0000 (08:52 -0700)]
add reviewer (#3755)

5 years agoImprove graph tuner dealing with Tuple (#3649)
Yao Wang [Sun, 11 Aug 2019 16:36:13 +0000 (09:36 -0700)]
Improve graph tuner dealing with Tuple (#3649)

* Improve graph tuner dealing with Tuple

* Add test case

* Move some data out of _base.py

* Fix lint

5 years ago[TOPI] Update tophub according to the fix in schedule (opencl and rocm) (#3752)
Lianmin Zheng [Sun, 11 Aug 2019 13:59:09 +0000 (21:59 +0800)]
[TOPI] Update tophub according to the fix in schedule (opencl and rocm) (#3752)

5 years ago[Relay] Fix Partial Evaluator, Add stricter checking for CheckWellFormed (#3749)
雾雨魔理沙 [Sun, 11 Aug 2019 01:23:23 +0000 (18:23 -0700)]
[Relay] Fix Partial Evaluator, Add stricter checking for CheckWellFormed (#3749)

* aot

* save

* save

* fix test

* remove vta changes

* lint

5 years agouse pip3 for python3 (#3742)
Henry [Fri, 9 Aug 2019 21:24:12 +0000 (14:24 -0700)]
use pip3 for python3 (#3742)

* use pip3 for python3

* make python3 as default

5 years ago[Relay] [Training] Fix ad for concatenate (#3729)
雾雨魔理沙 [Fri, 9 Aug 2019 19:40:16 +0000 (12:40 -0700)]
[Relay] [Training] Fix ad for concatenate (#3729)

* reproduce error

* fix

* lint

* lint

5 years ago Fix typo in ir_pass.h (#3741)
雾雨魔理沙 [Fri, 9 Aug 2019 08:51:20 +0000 (01:51 -0700)]
 Fix typo in ir_pass.h  (#3741)

5 years ago[VTA] [Chisel] Bug fix for VME Shell (#3737)
Benjamin Tu [Thu, 8 Aug 2019 20:53:27 +0000 (13:53 -0700)]
[VTA] [Chisel] Bug fix for VME Shell (#3737)

* fix

* fixes

5 years ago[CI] Update docker image ci_cpu,i386 to include verilator (#3738)
Tianqi Chen [Thu, 8 Aug 2019 20:10:23 +0000 (13:10 -0700)]
[CI] Update docker image ci_cpu,i386 to include verilator (#3738)

5 years ago[QNN] Requantize operator (#3531)
Animesh Jain [Thu, 8 Aug 2019 18:41:24 +0000 (11:41 -0700)]
[QNN] Requantize operator (#3531)

* [Relay] [Quantization] WIP - Common files for the qauntization work.

* [Relay] [Quantization] WIP - Prototyping requantize op.

* Requantize operator implementation.

Requantize converts one quantized tensor representation to another quantized
representation. The PR has following implementation features

- Requantize operator defined in qnn namespace - relay.qnn.requantize
- Lowering of the requantize to exisiting Relay operators
- Integer fixed point implementation of requantize
    - Two rounding modes - FE_UPWARDS (round towards infinity) and
    FE_AWAY_FROM_ZERO (std::round behavior)
- Floating point implementation as well, that can act as reference or can be
used for devices when FP32 computation is not used.
- Unit test cases

Relevant Issue - https://github.com/dmlc/tvm/issues/2351

Credit to TFLite and GemmLowp to provide reference implementations.

* Typo and lint fixes.

* Doc fix.

* Uncommenting the lint script (fixing mistake).

* Modifying the unit tests.

* Moving C++ files into src/relay/qnn

* Moving python files to python/tvm/relay/qnn. Some minor fixes.

* Moving the attrs.h inside the include directory.

* Pushing files that I forgot earlier. Changing util location.

* Incorporating comments. API change. Lint fixes.

* Modifying the GetFixedPointMultiplierShift API as per comments.

* Forgot the dialect change.

* Changing rewrite to qnn_lower.

* Renaming Quantize to Qnn for clarity.

* Remove use_int_domain.

* Incorportaing review comments.

* Adding API doc for QNN dialect.

* Move the qnn_lower pass to transform namespace.

* Moving from expr to module. Adding namespace in C++.

* Minor sentence rewrites. Added qnn namespace.

* Added the API doc.

* Chanding default out_dtype to int8. Adding a test with in/out_dtype as uint8.

* Style fixes. Better error messages.

* Adding documentation.

* More documentation fixes.

* Adding out dtype check for requantize.

* Adding corner case for FP32 to fixed point conversion.

* Adding extra line.

* Documentation fix.

* Adding static inline.

* Incorporating jackwish comment. Removed idtype from requantize lowering.

* Removing Quantize/Dequantize code. Restricting Requantize to (u)int8/int32.

* Style fixes.

* Fix the docs.

* Move to Legalize API.

5 years ago[DOCKER] Fix missing apt https transport support (#3735)
Marcus Shawcroft [Thu, 8 Aug 2019 17:36:36 +0000 (18:36 +0100)]
[DOCKER] Fix missing apt https transport support (#3735)

* [DOCKER] Fix missing apt https transport support

* [DOCKER] Drop superflous explicit sudo's

5 years agoRemove sccache from Rust install (#3728)
Nick Hynes [Wed, 7 Aug 2019 19:51:48 +0000 (12:51 -0700)]
Remove sccache from Rust install (#3728)

5 years agoTutorial: Build a Graph Convolutional Network on TVM (#3681)
Yulun Yao [Wed, 7 Aug 2019 16:59:00 +0000 (09:59 -0700)]
Tutorial: Build a Graph Convolutional Network on TVM (#3681)

* add build gcn tutorial

* add dgl to docker file

* add dgl to docker file

* Apply suggestions from code review

Co-Authored-By: 雾雨魔理沙 <lolisa@marisa.moe>
* add dgl to docker file

* rerun checks

* Revert "add build gcn tutorial"

This reverts commit dbe8b5f0e02a13fdd586a9faa58fd1326653afb0.

* resolve git issue

* resolve git issue

* resolve git issue

* apply marisa's comment

5 years ago[VTA][Dockerfile] Chisel dependencies for TSIM CI (#3721)
Thierry Moreau [Wed, 7 Aug 2019 15:53:41 +0000 (08:53 -0700)]
[VTA][Dockerfile] Chisel dependencies for TSIM CI (#3721)

5 years agoTake zero extent loops as NoOp and remove it and add unittest for the same (#3724)
Umang Yadav [Wed, 7 Aug 2019 15:51:18 +0000 (11:51 -0400)]
Take zero extent loops as NoOp and remove it and add unittest for the same (#3724)

5 years ago[Relay/TOPI][Op] Add variance and layer norm op (#3700)
Haichen Shen [Wed, 7 Aug 2019 11:34:53 +0000 (04:34 -0700)]
[Relay/TOPI][Op] Add variance and layer norm op (#3700)

* Add LayerNorm op

* update

* fix

* Add mean_std and mean_variance

* add std and update doc

* add license

* x

* lint

* x

* fix

* fix doc

5 years ago[Frontend][MXNet] Fix mxnet converter for hybridblock and add div_sqrt_dim (#3701)
Haichen Shen [Wed, 7 Aug 2019 04:27:06 +0000 (21:27 -0700)]
[Frontend][MXNet] Fix mxnet converter for hybridblock and add div_sqrt_dim (#3701)

* Fix mxnet converter for hybrid block

* tweak

* fix rebase

* fix

* add test

5 years agofix name (#3719)
雾雨魔理沙 [Wed, 7 Aug 2019 02:39:09 +0000 (19:39 -0700)]
fix name (#3719)

5 years ago[Relay] Legalize pass (#3672)
Animesh Jain [Tue, 6 Aug 2019 22:23:41 +0000 (15:23 -0700)]
[Relay] Legalize pass (#3672)

* [Relay] Rewrite pass.

This pass transforms an expression to other expression.

This pass has many usecases
 * Replace a expr to another expr, if the other expr has faster performance.
 * For ASICs, we might want to modify the inputs to adapt to the HW support.
 * Alter op layout can work in conjunction with this pass.

The supporting usecase is the Intel i8 x i8 conv. Intel HW supports u8 x i8 conv
in HW. Using this pass, we can replace an i8 x i8 conv to a sequence of
operators where one of the operators is now u8 x i8 conv. This will also help
automatic quantizaion performance.

* Better API name.

* Removing the conv2d legalization for x86. Will send a separate PR.

* Test name changes.

* Registering one funtion to register FTVMLegalize.

* Better comments.

5 years agoFix (2/2) [TOPI] conv2d schedule code (#3648) (#3717)
mingwayzhang [Tue, 6 Aug 2019 22:20:08 +0000 (15:20 -0700)]
Fix (2/2) [TOPI] conv2d schedule code (#3648) (#3717)

* Fix the tile_rx and tile_ry issue.

    Note that this patch depends on pull request #9 in tvm-distro.

5 years agoUpdate dmlc-core to the latest commit (#3716)
Krzysztof Parzyszek [Tue, 6 Aug 2019 21:58:24 +0000 (16:58 -0500)]
Update dmlc-core to the latest commit (#3716)

This includes changes to build TVM runtime for Hexagon.

5 years ago[relay][frontend] clean up tf frontend (#3710)
Zhi [Tue, 6 Aug 2019 21:05:06 +0000 (14:05 -0700)]
[relay][frontend] clean up tf frontend (#3710)

* clean up tf frontend

* fix get_relay_op

5 years agosafe to remove thread related headers? (#3713)
Liangfu Chen [Tue, 6 Aug 2019 20:58:38 +0000 (04:58 +0800)]
safe to remove thread related headers? (#3713)

5 years ago[Bugfix] Fix the issue that function pass modifies original module (#3712)
Haichen Shen [Tue, 6 Aug 2019 19:25:59 +0000 (12:25 -0700)]
[Bugfix] Fix the issue that function pass modifies original module (#3712)

* fix

* fix interpreter

5 years ago[Relay] [TOPI] `{relay,topi}.nn.sparse_transpose` for **Square** CSR matrices (#3707)
Yulun Yao [Tue, 6 Aug 2019 01:13:22 +0000 (18:13 -0700)]
[Relay] [TOPI] `{relay,topi}.nn.sparse_transpose` for **Square** CSR matrices (#3707)

* add build gcn tutorial

* add transpose operator for square sparse matrices

* remove extra files

* change loop tag

* comply with lint

* comply with lint -- line too long

* comply with lint

* lint check

* lint check

* lint check

* apply marisa and theirry's reviews

5 years agoExport tvm::relay::OpRegistry::OpRegistry (#3711)
Junru Shao [Mon, 5 Aug 2019 22:16:14 +0000 (15:16 -0700)]
Export tvm::relay::OpRegistry::OpRegistry (#3711)

5 years ago[CI] Update GPU docker (#3709)
Tianqi Chen [Mon, 5 Aug 2019 21:55:25 +0000 (14:55 -0700)]
[CI] Update GPU docker (#3709)

5 years agoQuit and clean when TVM is interrupted (#3640)
ghostplant [Mon, 5 Aug 2019 16:31:55 +0000 (00:31 +0800)]
Quit and clean when TVM is interrupted (#3640)

5 years agoMetal reinterpret fix (#3706)
Andrew Tulloch [Mon, 5 Aug 2019 16:31:19 +0000 (09:31 -0700)]
Metal reinterpret fix (#3706)

5 years ago[Relay] Partial Evaluator do concatenate, and has better termination checker for...
雾雨魔理沙 [Mon, 5 Aug 2019 16:23:36 +0000 (09:23 -0700)]
[Relay] Partial Evaluator do concatenate, and has better termination checker for scalar. (#3703)

* save

lint some

lint

lint

add charrnn

save

save

save

remove debug

remove debug

remove space

refactor

save

rewrite dce

* reset files

* join -> meet

* lint

* address review comment

* wordsmith

5 years ago[TOPI] Update softmax compute and CPU schedule (#3680)
Jon Soifer [Mon, 5 Aug 2019 02:46:28 +0000 (19:46 -0700)]
[TOPI] Update softmax compute and CPU schedule (#3680)

* Update Softmax compute and CPU schedule

* Add C++ compute

* Fix schedule

* Update CUDA and OpenGL schedules

* Fix log_softmax

* Fix hls and opengl schedules

* Fix CUDA schedule

5 years agoFix gather_nd in Relay (#3442)
Huilin Qu [Sat, 3 Aug 2019 23:55:22 +0000 (19:55 -0400)]
Fix gather_nd in Relay (#3442)

* Fix gather_nd in Relay

* Add test cases for gather_nd.

5 years ago[VTA] [Chisel] Added Chisel Module Unit Test Infrastructure (#3698)
Benjamin Tu [Sat, 3 Aug 2019 06:04:38 +0000 (23:04 -0700)]
[VTA] [Chisel] Added Chisel Module Unit Test Infrastructure (#3698)

* added wholething

* changed build and makefile

5 years agoAdd an option to build with -pthread (ON by default) (#3671)
abergeron [Sat, 3 Aug 2019 04:09:44 +0000 (00:09 -0400)]
Add an option to build with -pthread (ON by default) (#3671)

5 years ago[Relay] [Error] Fix error in partial evaluator (#3693)
雾雨魔理沙 [Fri, 2 Aug 2019 17:35:27 +0000 (10:35 -0700)]
[Relay] [Error] Fix error in partial evaluator (#3693)

* fix

* lint

5 years ago[AutoTVM] Fix hang/crash issues on feature extraction (#3689)
Lianmin Zheng [Fri, 2 Aug 2019 16:14:27 +0000 (00:14 +0800)]
[AutoTVM] Fix hang/crash issues on feature extraction (#3689)

* [AutoTVM] Fix hang/crash issues on feature extraction

* Update xgboost_cost_model.py

* fix lint

5 years agoAlign the naming rule for OpAttributeUnImplemented (#3695)
Neo Chien [Fri, 2 Aug 2019 15:52:00 +0000 (23:52 +0800)]
Align the naming rule for OpAttributeUnImplemented (#3695)

5 years ago[DOCKER] Add DGL to {ci_gpu, demo_cpu, demo_gpu} docker images (#3692)
Yulun Yao [Fri, 2 Aug 2019 15:51:14 +0000 (08:51 -0700)]
[DOCKER] Add DGL to {ci_gpu, demo_cpu, demo_gpu} docker images (#3692)

* add dgl to docker file

* add dgl to docker file

5 years ago[TOPI] Memoize winograd matrix (#3687)
Lianmin Zheng [Fri, 2 Aug 2019 15:50:33 +0000 (23:50 +0800)]
[TOPI] Memoize winograd matrix (#3687)

* [TOPI] Memoize winograd matrix

* lint

* Fix name

5 years ago[Relay][Quantization] KL-divergence-based per-layer calibration (#3538)
Wuwei Lin [Fri, 2 Aug 2019 03:55:27 +0000 (20:55 -0700)]
[Relay][Quantization] KL-divergence-based per-layer calibration (#3538)

* [Relay][Quantization] Support floating-point scale

* [Relay][Quantization] KL-divergence calibration on dataset

* Fix unhandled LeftShift case in QuantizeRealize

* Fix lint

* drop QBias

* fix lint

* address comments

* address comments

* Update comments

* address comments

* lint

* kQIdentity = 0

5 years ago[Relay][VM] Support execution on devices (#3678)
Wei Chen [Thu, 1 Aug 2019 21:47:11 +0000 (14:47 -0700)]
[Relay][VM] Support execution on devices (#3678)

* [Relay][VM] Support execution on devices

* Reduce Copy calls

* Cleanup

* Lint

* CR comments

* Merge test into test_vm.py

5 years agoAdd shuffle support to TVM (#3633)
Jian Weng [Thu, 1 Aug 2019 19:52:33 +0000 (12:52 -0700)]
Add shuffle support to TVM (#3633)

5 years agoEnable the sparse schedule (#3651)
sf-wind [Thu, 1 Aug 2019 19:49:40 +0000 (12:49 -0700)]
Enable the sparse schedule (#3651)

5 years agoAdd support for Tensorflow operators log1p, cos, sin (#3614)
alexgl-github [Thu, 1 Aug 2019 19:46:39 +0000 (12:46 -0700)]
Add support for Tensorflow operators log1p, cos, sin (#3614)

The patch adds support for Tensorflow operators log1p and cos
Tensorflow log1p is described at https://www.tensorflow.org/api_docs/python/tf/math/log1p
Tensorflow cos is described at https://www.tensorflow.org/api_docs/python/tf/math/cos
Tensorflow sin is described at https://www.tensorflow.org/api_docs/python/tf/math/sin

5 years ago[Relay] Strict mode in pattern matching (#3620)
雾雨魔理沙 [Thu, 1 Aug 2019 18:52:13 +0000 (11:52 -0700)]
[Relay] Strict mode in pattern matching (#3620)

* add fatal

lint

lint

lint

do

make completeness check an error

lint

remove fatal

* fix test

* reset parser file

* remove unneeded import

* Update python/tvm/relay/adt.py

Co-Authored-By: Steven S. Lyubomirsky <slyubomirsky@gmail.com>
* Update include/tvm/relay/adt.h

Co-Authored-By: Steven S. Lyubomirsky <slyubomirsky@gmail.com>
* Eliminate trailing whitespace (my fault)

5 years ago[Relay][Frontend] Fix typo names in frontend (#3685)
Yifan Xiong [Thu, 1 Aug 2019 16:46:23 +0000 (00:46 +0800)]
[Relay][Frontend] Fix typo names in frontend (#3685)

Fix typo names in caffe2 and onnx frontend:
* sotrage_order -> storage_order
* OpNotInplemented -> OpNotImplemented

5 years agoMake tests multi-process friendly. (#3683)
Tim Hatch [Thu, 1 Aug 2019 16:27:58 +0000 (09:27 -0700)]
Make tests multi-process friendly. (#3683)

This side effect at module import time has a race condition between the "exists" check and the "mkdir" call.  The safer thing is to just call mkdir and catch the "already exists" error which is what makedirs does.

5 years agoReplace learnt with learned (#3684)
Alexander Pivovarov [Thu, 1 Aug 2019 15:31:09 +0000 (08:31 -0700)]
Replace learnt with learned (#3684)

5 years ago[DOC] Update ssd doc to avoid confusion. (#3677)
Leyuan Wang [Wed, 31 Jul 2019 20:45:58 +0000 (13:45 -0700)]
[DOC] Update ssd doc to avoid confusion. (#3677)

* intel graphics conv2d bugs fixed for inception_v3

* intel conv2d api updated, nn input size 4 condition added

* review addressed

* move conv_tags to attributes

* ssd doc updated

* address comment

5 years ago[Relay][VM] Relay VM serialization (#3647)
Zhi [Wed, 31 Jul 2019 16:02:15 +0000 (09:02 -0700)]
[Relay][VM] Relay VM serialization (#3647)

* relay vm serialization

* fix lint

* load params, fix stream

* lint

* fix typo

5 years ago[TEST] Comptiable with python3.5 (#3675)
lixiaoquan [Wed, 31 Jul 2019 15:37:54 +0000 (23:37 +0800)]
[TEST] Comptiable with python3.5 (#3675)

5 years ago[TOPI][CUDA] schedule for group_conv2d (#3663)
Wuwei Lin [Wed, 31 Jul 2019 08:26:05 +0000 (16:26 +0800)]
[TOPI][CUDA] schedule for group_conv2d (#3663)

* [TOPI][CUDA] schedule for group_conv2d

* Fix #flops

5 years ago[VTA] VTA Compilation Script for Intel FPGA (#3494)
Liangfu Chen [Wed, 31 Jul 2019 07:19:54 +0000 (15:19 +0800)]
[VTA] VTA Compilation Script for Intel FPGA (#3494)

* initial compilation script for chisel-vta;

* replace tabs with spaces;

* compile script for de10-nano;

* remove generated verilog source code;

* remove `altsource_probe`, `debounce`, `edge_detect` ip;

* replace quartus project files with a single tcl script;

* Update install.md

* improved makefile-based compilation script;

* complete makefile-based compilation of chisel-vta for de10-nano;

* install quartus;

* conversion to .rbf file;

* document chisel-vta compilation process for de10-nano;

* rename generated bitstream file;

* download and extract custom ip for de10-nano;

* minor change

* minor change

* fix indentation;

* bug fix;

* improved robustness in makefile;

* clean up;

* add `.sdc .ipx .qsys` allowance in jenkins;

* add ASF header;

* add ASF header;

* remove IntelShell.scala, update vta_hw.tcl, clean up Makefile & soc_system.qsys;

* add ASF header;

* keep sources compact;

* keep sources compact;

* it's not necessary now

* AXI4LiteClient -> AXI3Client for IntelShell

* remove connection to fpga_only_master;

* a few important bug fix: wire reset pin, and set host_r_last to high

* remove intel specific interface definition;

* add NO_DSP option in Makefile;

* AXI4Lite is not used in IntelShell;

* minor fix: disable dsp and use logic instead;

* quartus version change: 18.0 -> 18.1

* remove altera related statement;

* compose compile_design.tcl

* initial tcl script for soc_system generation;

* remove .qsys file;

* remove unused;

* .qsys can be generated by tcl script;

* remove hps_io and shrink size of soc_system;

* integrate into makefile;

* version change: 18.0 -> 18.1

* add sample config file for de10-nano;

* parameterize DEVICE and PROJECT_NAME

* remove extra lines;

* brief description on flashing sd card image for de10-nano

* docs on building additional components

* parameterize DEVICE and DEVICE_FAMILY

* parameterize DEVICE and DEVICE_FAMILY

* parameterize DEVICE and DEVICE_FAMILY

* de10-nano -> de10nano

* minor change

* add comment in code and document in order to address review comments;

5 years agoAdd yolov3-tiny to the tutorial. (#3674)
Balint Cristian [Wed, 31 Jul 2019 07:10:16 +0000 (10:10 +0300)]
Add yolov3-tiny to the tutorial. (#3674)

5 years agoadd reviewer - slyubomirsky (#3673)
Haichen Shen [Wed, 31 Jul 2019 01:22:51 +0000 (18:22 -0700)]
add reviewer - slyubomirsky (#3673)

5 years ago[RPC] Terminate worker's childs first. (#3669)
Balint Cristian [Tue, 30 Jul 2019 22:06:50 +0000 (01:06 +0300)]
[RPC] Terminate worker's childs first. (#3669)

5 years ago[VTA] Support for batched inference (#3661)
Thierry Moreau [Tue, 30 Jul 2019 21:01:31 +0000 (14:01 -0700)]
[VTA] Support for batched inference (#3661)

* fix in IR pass to support padding on 6-d tensors

* support for both N>1 and N==1 for padding

* batch size > 1 tuning and base config

* output formatting

* batch conv2d

* print all category results

* revert to single-batch config

* pick record best

* fix conv test

* improving reporting

* address batching bug in fast simulator

* fix

5 years agoremoving deprecated script (#3667)
Thierry Moreau [Tue, 30 Jul 2019 21:00:38 +0000 (14:00 -0700)]
removing deprecated script (#3667)

5 years ago[TOPI] Enable standalone wheel build (#3657)
Josh Fromm [Tue, 30 Jul 2019 16:29:56 +0000 (09:29 -0700)]
[TOPI] Enable standalone wheel build (#3657)

* Fixed topi bdist_wheel build to include libraries.

* Removed unneeded imports

5 years ago[TOPI] Fix traverse function not inline zero-input op (#3623)
Wuwei Lin [Tue, 30 Jul 2019 15:25:15 +0000 (23:25 +0800)]
[TOPI] Fix traverse function not inline zero-input op (#3623)

* Fix traverse_inline not inline zero input op properly

* Add where to python and set tag to broadcast

* Fix inline

* test

* fix test target

* fix

5 years agoROCm: Add SaveToFile and LoadFile (#3665)
Thomas Viehmann [Tue, 30 Jul 2019 14:54:16 +0000 (16:54 +0200)]
ROCm: Add SaveToFile and LoadFile (#3665)

...and add rocm module_save to the tests.

5 years agotvm/contrib/rocm: improve finding of ld.lld (#3664)
Thomas Viehmann [Tue, 30 Jul 2019 10:40:50 +0000 (12:40 +0200)]
tvm/contrib/rocm: improve finding of ld.lld (#3664)

This refines the detection of ld.lld matching the neighbouring clang
file. This is particularly helpful on Ubuntu/Debian when either the
default ld.lld is not installed or the versioned one is preferable for
consistency.

@tqchen I think you last touched the clang equivalent in #3590 .

5 years agoPrint llvm source by default in ROCMModuleNode::GetSource (#3662)
Thomas Viehmann [Tue, 30 Jul 2019 09:30:46 +0000 (11:30 +0200)]
Print llvm source by default in ROCMModuleNode::GetSource (#3662)