GitLab

* initial testing

* change tolerance

* remove extra changes
Co-authored-by: Shucai Xiao <shucai.xiao@amd.com>
Co-authored-by: Paul Fultz II <pfultz2@yahoo.com>
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>

* Unify the vectorized and non-vectorized path

* Formatting

* Make fusion easily extendable

* Add skip layernorm fusion

* Formatting

* Call correct layernorm function

* Fix compile errors

* Add DCE

* Add test for skip layernorm

* Formatting

* Remove unused typedef

* Formatting

* Fix tidy issues

* Formatting
Co-authored-by: Shucai Xiao <shucai.xiao@amd.com>

* code backup

* clang format

* change corresponding tool files

* clang format
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>

* Fix fusions in bert model

* Formatting

* Add unit tests

* Formatting

* Fix one_half matcher

* Workaround ICE on gcc

* Formatting

* Tidy fixes
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>

* add flag

* formatting

* remove env variable

* fix api expression

* add api test

* add api test

* add op test

* formatting

* fix function name

* fix syntax

* formatting

* modify test

* remove test and update doc

* move test to new file

* formatting

* revert test files

* rewrite check

* New
Co-authored-by: Paul Fultz II <pfultz2@yahoo.com>

* Fuse gemm in fuse ops

* Formatting

* Add const ref

* Remove assert

* Skip already fused gemms

* Skip already fused gemm

* Formatting

* Use float_equal

* Avoid non-standard shapes for inputs

* Formatting
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>

* Use increment instead of division to compute register offset

* Formatting

* Limit layernorm to 1024 elements

* Formatting

* Add verification to driver

* Formatting

* Remove early return

* Use block_size 256

* Vectorize the kernel

* Formatting

* Convert to vector type

* Add layernorm tests

* Formatting

* Formatting

* Refactor layernorm to run both algos

* Formatting

* Fix compile error

* Fix tidy warnings

* Formatting

* Add layernorm function

* Formatting

Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>

* move initialization of miopen fusion operators to finalize method

* clang format

* fix cppcheck error

* clang format

* fix review comments

* clang format

* removed an unnecessary assert

* Register ops for main migraphx

* Formatting

* Register cpu ops

* Formatting

* Show list of operators in the driver

* Formatting

* Simplify regiter

* Try to register gpu ops

* Fix compiler errors

* Register rest of the gpu operators

* Add some tests

* Formatting

* Fix gcc compiler warnings

* Formatting

* Fix tidy warnings

* Fix compile error

* Use correct op name

* Register layer norm

* Use const ref

* Make run const

* fix pad calc

* bert tf passes correctness

* formatting

* add test

* formatting

* remove comment

* add inline

* formatting

* fix order for literal

* formatting

* test no mul_add

* formatting

* debug layernorm

* debug layernorm

* manual merge

* more progress

* formatting

* remove miopen batchnorm

* remove headers

* Fix compile error with no dpp reductions

* fix indices

* formatting

* change matcher

* formatting

* remove binds

* formatting

* disable tf matcher

* formatting

* use fast div

* formatting

* fix matcher

* formatting

* remove comment

* move find_matches

* add assert

* formatting

* fix deepcode issue
Co-authored-by: Paul <pfultz2@yahoo.com>
Co-authored-by: Shucai Xiao <shucai.xiao@amd.com>
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>

* Add reduce dims

* Formatting

* Reduce dims on the gpu

* Formatting

* Fix tidy issues

* Convert to assert

* Reduce dims for contiguous

* Formatting

* Remove move

* Fix arguments used

* Formatting

* Fix warnings

* Formatting
Co-authored-by: Shucai Xiao <shucai.xiao@amd.com>

* initial progress

* formatting

* add pooling changes

* formatting

* change eliminate_pad

* formatting

* rename var

* fomratting

* update op shape test and compute

* formatting

* revert conv constructor

* formatting

* change initializer

* formatting

* fix tidy

* change quant conv and shape check

* add tests and fixes

* formatting

* fix type

* fix conv test

* formatting

* add pooling and bn tests

* formatting

* add inconsistent attr tests

* fix padding issue

* formatting

* fix review comments, remove duplicate test

* formatting

* fix variable

* fix assert bug

* fix attr check

* remove std
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>

* fix pad calc

* Add decompose pass

* Add decompose test

* Formatting

* bert tf passes correctness

* formatting

* Add remap

* Formatting

* add test

* formatting

* remove comment

* Add compute method for dot

* Formatting

* add inline

* Add finder for horizontal fusion

* Formatting

* Formatting

* Reuse predicate

* formatting

* fix order for literal

* formatting

* add test for gelu

* formatting

* added add_gelu fusion

* Add gemm fusions

* Formatting

* add files

* formatting

* test no mul_add

* formatting

* progress on div

* formatting

* continue work on pass

* remove layernorm opt

* revert reduce file

* Add some fixes for convolution

* Formatting

* Fix shape tests

* Formatting

* Reuse axis equal

* Add initial split fusion

* Formatting

* Update offset

* Workaround outputs that cant accept nonstandard shapes

* Formatting

* Add check for split concat

* Formatting

* Add missing headers

* Formatting

* Add tests

* Formatting

* add optimization for bert

* code backup for bert optimization

* continue testing

* formatting

* fix matcher

* formatting

* add gelu_fn and tests

* formatting

* fix matcher, remove extra tests

* formatting

* fix matcher

* add missing files

* add find_layernorm

* add add_transpose to cmake file

* code backup for the contigous fusion

* refine ops fusion

* clang format

* fixed bug in previous optimization

* clang format

* add more optimization

* remove unnecessary code

* refinement of the fustion code

* clang format

* fixed a bug

* add used_once

* formatting

* start on new gelu

* formatting

* add matchers in fuse_ops

* formatting

* add dce to fix add_gelu

* add simplify_rsqrt and test

* formatting

* debugging value for matcher

* formatting

* add more to matchers

* formatting

* fix errors

* remove onnx gen

* add any_arg, change matchers to use either_arg

* formatting

* clang format

* formatting

* add used_once

* formatting

* code cleanup

* clang format

* fixed a bug

* remove unnecessary code

* refine comments

* optimize bert to remove more contiguous

* clang format

* remove unnecessary code

* add unit tests for bert optimization

* clang format

* fix review comments

* clang format

* refine a fusion of reshape and slice

* clang format

* fix cppcheck error

* fix review comments

* add the fusion of slice and transpose

* clang format

* add another optimization to fuse slice and transpose

* clang format

* fix review comments

* clang format

* fix review comments

* clang format

* fix review comments
Co-authored-by: Khalique <15948690+kahmed10@users.noreply.github.com>
Co-authored-by: Paul <pfultz2@yahoo.com>
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>
Co-authored-by: Shucai Xiao <scxiao@prj47-rack-99.local.lan>

* fix pad calc

* bert tf passes correctness

* formatting

* add test

* formatting

* remove comment

* add inline

* formatting

* fix order for literal

* formatting

* add test for gelu

* formatting

* added add_gelu fusion

* add files

* formatting

* remove layernorm opt

* revert reduce file

* add gelu_fn and tests

* formatting

* fix matcher, remove extra tests

* formatting

* fix matcher

* add used_once

* formatting

* start on new gelu

* formatting

* add matchers in fuse_ops

* formatting

* add dce to fix add_gelu

* add simplify_rsqrt and test

* formatting

* debugging value for matcher

* formatting

* add more to matchers

* formatting

* fix errors

* remove onnx gen

* add any_arg, change matchers to use either_arg

* formatting

* formatting

* add used_once

* formatting
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>

* Add decompose pass

* Add decompose test

* Formatting

* Add remap

* Formatting

* Add compute method for dot

* Formatting

* Add finder for horizontal fusion

* Formatting

* Formatting

* Reuse predicate

* Add gemm fusions

* Formatting

* Add some fixes for convolution

* Formatting

* Fix shape tests

* Formatting

* Reuse axis equal

* Add initial split fusion

* Formatting

* Update offset

* Workaround outputs that cant accept nonstandard shapes

* Formatting

* Add check for split concat

* Formatting

* Add missing headers

* Formatting

* Add tests

* Formatting

* Add more testing

* Formatting

* Fix when there is duplicate splits in inputs

* Formatting

* Fix mismatch iterators

* Add tests for dot fusions

* Formatting

* Add test for convolution

* Formatting

* Fix tidy issues

* Add more tests

* Formatting

* Ignore build directory for codecov

* Add test for groups

* Formatting

* Add more tests for groups

* Formatting

* Add test for missing end slice

* Add newline

* Remove unused function

* Add support for when beta is not 1

* Formatting

* Add test for scalar

* Add one more scalar test
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>

* fix pad calc

* modify clip for more args

* formatting

* add test, flip order, revert to unary

* fix error msg

* add min and max args to clip

* formatting

* fixes to quantization

* formatting

* fix logic and add extra test

* formatting

* fix logic, add extra test

* formatting

* fix bug in test
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>
Co-authored-by: Paul Fultz II <pfultz2@yahoo.com>

* improve unsqueeze to support negative axis and parsing scalar

* clang format

* add a test example for the negative axis of unsqueeze

* improve the squeeze operator to support negative axis

* clang format

* fixed a small bug in the lrn implementation

* clang format

* support negative axis in argmax and argmin

* clang format

* improve flatten to support negative axis

* clang format

* change softmax/logsoftmax to support negative axis

* clang format

* improve transpose by adding default perm

* clang format

* add one more dimens for tensor size

* add one more dimens for tensor size

* disable conv ops fusion for non-symmetric cases

* clang format

* fixed review comments

* move computing axis from the device function to the compute function

* clang format

* move computing axis from device function to the operator computing function

* clang format
Co-authored-by: mvermeulen <5479696+mvermeulen@users.noreply.github.com>

* Fix bug in bert accuraccy

* Formatting

* add another test

* Fix add and overflow

* Formatting

* Fix bug in shape_for_each

* Use front instead of iterator

* Use result.front()

* Split add_unary files

* Formatting

* Fix incorrect last index

* Remove comment

* Inline function

* Fix carry check

* Fix metadata errors

* Formatting

* Reflow

* Reflow

* initial testing of add_clip fusion

* formatting

* clipped relu fusion

* formatting

* remove some executables, add fusion test

* formatting

* remove clipped_relu code

* fix clang-tidy

* revert changes to cmake files

* remove fusion from weight map

* formatting

* fix syntax error

* formatting

* fix syntax error

* fix syntax error

* formatting

* Disable fusion when winograd is used except for 3x3

* Formatting

* add tests, fix bug in ternary op

* formatting

* uncomment fusion