* first commit
* turned C from input to constant; force C constant in impl; better handling 0d/1d cases
* integrate with gemm from ficus nn
* fix const inputs
* adjust threshold for int8 tryQuantize
* adjust threshold for int8 quantized 2
* support batched gemm and matmul; tune threshold for rcnn_ilsvrc13; update googlenet
* add gemm perf against innerproduct
* add perf tests for innerproduct with bias
* fix perf
* add memset
* renamings for next step
* add dedicated perf gemm
* add innerproduct in perf_gemm
* remove gemm and innerproduct perf tests from perf_layer
* add perf cases for vit sizes; prepack constants
* remove batched gemm; fix wrong trans; optimize KC
* remove prepacking for const A; several fixes for const B prepacking
* add todos and gemm expression
* add optimized branch for avx/avx2
* trigger build
* update macros and signature
* update signature
* fix macro
* fix bugs for neon aarch64 & x64
* add backends: cuda, cann, inf_ngraph and vkcom
* fix cuda backend
* test commit for cuda
* test cuda backend
* remove debug message from cuda backend
* use cpu dispatcher
* fix neon macro undef in dispatcher
* fix dispatcher
* fix inner kernel for neon aarch64
* fix compiling issue on armv7; try fixing accuracy issue on other platforms
* broadcast C with beta multiplied; improve func namings
* fix bug for avx and avx2
* put all platform-specific kernels in dispatcher
* fix typos
* attempt to fix compile issues on x64
* run old gemm when neon, avx, avx2 are all not available; add kernel for armv7 neon
* fix typo
* quick fix: add macros for pack4
* quick fix: use vmlaq_f32 for armv7
* quick fix for missing macro of fast gemm pack f32 4
* disable conformance tests when optimized branches are not supported
* disable perf tests when optimized branches are not supported
* decouple cv_try_neon and cv_neon_aarch64
* drop googlenet_2023; add fastGemmBatched
* fix step in fastGemmBatched
* cpu: fix initialization ofb; gpu: support batch
* quick followup fix for cuda
* add default kernels
* quick followup fix to avoid macro redef
* optmized kernels for lasx
* resolve mis-alignment; remove comments
* tune performance for x64 platform
* tune performance for neon aarch64
* tune for armv7
* comment time consuming tests
* quick follow-up fix
OCL_FP16 MatMul with large batch
* Workaround FP16 MatMul with large batch
* Fix OCL reinitialization
* Higher thresholds for INT8 quantization
* Try fix gemm_buffer_NT for half (columns)
* Fix GEMM by rows
* Add batch dimension to InnerProduct layer test
* Fix Test_ONNX_conformance.Layer_Test/test_basic_conv_with_padding
* Batch 16
* Replace all vload4
* Version suffix for MobileNetSSD_deploy Caffe model
Resolve uncovered CUDA dnn layer #24080
### Pull Request Readiness Checklist
* Gelu activation layer on CUDA
* Try to relax GEMM from ONNX
resolves https://github.com/opencv/opencv/issues/24064
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch
- [x] There is a reference to the original bug report and related work
- [x] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
Patch to opencv_extra has the same branch name.
- [x] The feature is well documented and sample code can be built with the project CMake
DNN: optimize the speed of general Depth-wise #23952
Try to solve the issue: https://github.com/opencv/opencv/issues/23941
### Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch
- [ ] There is a reference to the original bug report and related work
- [ ] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
Patch to opencv_extra has the same branch name.
- [ ] The feature is well documented and sample code can be built with the project CMake
dnn: add layer normalization for vision transformers
* add layer norm onnx parser, impl and tests
* add onnx graph simplifier for layer norm expanded
* handle the case when constants are of type Initializer
* add test case for layer norm expanded with initializers
* use CV_Assert & CV_CheckType in place of CV_Assert_N; use forward_fallback for OCL_FP16
* use const ref / ref in parameters of invoker::run; extract inner const if from nested loop; use size_t in place of ull
* template hasBias
* remove trailing whitespace
* use pointer parameter with null check; move normSize division & mean_square division outside of loop; use std::max to ensure positive value before std::sqrt
* refactor implementation, optimize parallel_for
* disable layer norm expanded
* remove the removal of layer norm optional outputs
DNN: reduce the memory used in convolution layer
* reduce the memory in winograd and disabel the test when usage memory is larger than 2gb.
* remove VERY_LOG tag
Reimplementation of Element-wise layers with broadcasting support
* init
* semi-working initial version
* add small_vector
* wip
* remove smallvec
* add nary function
* replace auto with Mat in lambda expr used in transform
* uncomment asserts
* autobuffer shape_buf & step_buf
* fix a missing bracket
* fixed a missing addLayer in parseElementWise
* solve one-dimensional broadcast
* remove pre_broadcast_transform for the case of two constants; fix missing constBlobsExtraInfo when addConstant is called
* one autobuffer for step & shape
* temporal fix for the missing original dimension information
* fix parseUnsqueeze when it gets a 1d tensor constant
* support sum/mean/min/max with only one input
* reuse old code to handle cases of two non-constant inputs
* add condition to handle div & mul of two non-constant inputs
* use || instead of or
* remove trainling spaces
* enlarge buf in binary_forward to contain other buffer
* use autobuffer in nary_forward
* generate data randomly and add more cases for perf
* add op and, or & xor
* update perf_dnn
* remove some comments
* remove legacy; add two ONNX conformance tests in filter
* move from cpu_denylist to all_denylist
* adjust parsing for inputs>=2
Co-authored-by: fengyuentau <yuantao.feng@opencv.org.cn>
* dnn: LSTM optimisation
This uses the AVX-optimised fastGEMM1T for matrix multiplications where available, instead of the standard cv::gemm.
fastGEMM1T is already used by the fully-connected layer. This commit involves two minor modifications:
- Use unaligned access. I don't believe this involves any performance hit in on modern CPUs (Nehalem and Bulldozer onwards) in the case where the address is actually aligned.
- Allow for weight matrices where the number of columns is not a multiple of 8.
I have not enabled AVX-512 as I don't have an AVX-512 CPU to test on.
* Fix warning about initialisation order
* Remove C++11 syntax
* Fix build when AVX(2) is not available
In this case the CV_TRY_X macros are defined to 0, rather than being undefined.
* Minor changes as requested:
- Don't check hardware support for AVX(2) when dispatch is disabled for these
- Add braces
* Fix out-of-bounds access in fully connected layer
The old tail handling in fastGEMM1T implicitly rounded vecsize up to the next multiple of 8, and the fully connected layer implements padding up to the next multiple of 8 to cope with this. The new tail handling does not round the vecsize upwards like this but it does require that the vecsize is at least 8. To adapt to the new tail handling, the fully connected layer now rounds vecsize itself at the same time as adding the padding(which makes more sense anyway).
This also means that the fully connected layer always passes a vecsize of at least 8 to fastGEMM1T, which fixes the out-of-bounds access problems.
* Improve tail mask handling
- Use static array for generating tail masks (as requested)
- Apply tail mask to the weights as well as the input vectors to prevent spurious propagation of NaNs/Infs
* Revert whitespace change
* Improve readability of conditions for using AVX
* dnn(lstm): minor coding style changes, replaced left aligned load
Add support for Conv1D on OpenCV backend
* Add support for Conv1D on OpenCV backend
* disable tests on other targets/backends
* Fix formatting
* Restore comment
* Remove unnecessary flag and fix test logic
* Fix perf test
* fix braces
* Fix indentation, assert check and remove unnecessary condition
* Remove unnecessary changes
* Add test cases for variable weights and bias
* dnn(conv): fallback on OpenCV+CPU instead of failures
* coding style
