Learning backend cuda

Context

This MR is intended to add the forward implementations for Accuracy computation operators:

• And
• ArgMax
• ReduceSum

And the forward implementations for operators used in the MSE #20 (closed):

• Mul
• ReduceMean
• Pow

Also a fix to #23 (closed) is added: update Tensor descriptor when shape changes.
Also a fix to #7 (closed) is added: force conv fwd algorithm.

We also add operator Ln forward implementation to support #24 (closed).

Modified files

• AndImpl.hpp, AndImpl_CUDA_kernels.hpp, AndImpl_CUDA_kernels.cu, AndImpl.cpp and Tests_AndImpl.cpp, add And forward impl;
• ArgMaxImpl.hpp, ArgMaxImpl_CUDA_kernels.hpp, ArgMaxImpl_CUDA_kernels.cu, ArgMaxImpl.cpp and Tests_ArgMaxImpl.cpp, add ArgMaxforward impl;
• ReduceSumImpl.hpp, ReduceSumImpl_CUDA_kernels.hpp, ReduceSumImpl_CUDA_kernels.cu, ReduceSumImpl.cpp and Tests_ReduceSumImpl.cpp, add ReduceSum forward impl;
• LnImpl.hpp, LnImpl_CUDA_kernels.hpp, LnImpl_CUDA_kernels.cu, LnImpl.cpp and Tests_LnImpl.cpp, add Ln forward impl;
• ConvImpl.hpp, ConvImpl.cpp and Tests_ConvImpl.cpp, force Conv Fwd algorithm;
• TensorImpl.hpp, update Tensor descriptor when tensor shape has changed;
• MulImpl.hpp, MulImpl.cpp and Test_MulImpl.cpp, add Mul forward() and backward();
• ReduceMeanImpl.hpp, ReduceMeanImpl.cpp and Test_ReduceMeanImpl.cpp, add ReduceMean forward() and backward();
• PowImpl.hpp, PowImpl.cpp and Test_PowImpl.cpp, add Pow forward() and backward();

TODO

• Mul
• ReduceMean
• Pow
• And
• ArgMax
• ReduceSum
• Fix TensorImpl

added 1 commit

• 3d2b9a2a - enhance ArgMax kernel

Compare with previous version

marked this merge request as ready

added 1 commit

• 5e665c77 - support keepDims and NoopWithEmptyAxes attr in Reduce Ops

Compare with previous version

added 1 commit

• 09aaf752 - update tensorDescriptor if dimensions change

Compare with previous version

added 2 commits

• 012caa3b - add Div forward
• a8ba69bc - add Ln forward Imp

Compare with previous version

added 4 commits

• a8ba69bc...2e91ef4a - 2 commits from branch eclipse/aidge:dev
• 1e036709 - handle Conv with no bias connected
• 6143460a - Merge branch 'dev' of gitlab.eclipse.org:eclipse/aidge/aidge_backend_cuda into...

Compare with previous version

assigned to @thibaultallenet

unassigned @thibaultallenet

requested review from @thibaultallenet

mentioned in merge request aidge_core!178 (merged)

added 2 commits

• 59d71428 - add test for updating tensor descriptor when shape changes
• 8c99d196 - force conv fwd algorithm

Compare with previous version

added 1 commit

• f41feb83 - force conv fwd algorithm

Compare with previous version

changed the description

mentioned in issue #7 (closed)

resolved all threads

added 1 commit

• f31672fe - template pow kernel

Compare with previous version

added 1 commit

• 78a6e1a7 - Fix ReduceMean call

Compare with previous version

changed the description

mentioned in merge request !31 (closed)

added 2 commits

• 106e2607 - Fix ReduceSum call
• 574eec00 - Fix Coverage error

Compare with previous version

added 1 commit

• 33e22727 - Fix Coverage error

Compare with previous version

added 1 commit

• 2592072a - Fix Coverage error

Compare with previous version

added 1 commit

• d63b7d3c - Fix Coverage error

Compare with previous version

added 1 commit

• e88ce115 - fix : wrong job was retrieved + coverage was enabled in optionnal C++ builds

Compare with previous version

added 1 commit

• 87245bdc - chore: upd ci files ref

Compare with previous version

added 1 commit

• a703ff30 - chore : upd ci files ref

Compare with previous version

added 1 commit

• 8fd1b0b2 - chore : upd ci files ref

Compare with previous version

added 1 commit

• aeb9d2a6 - fix : ci job names to retrieve

Compare with previous version

mentioned in merge request gitlab_shared_files!12 (merged)

added 1 commit

• d14bfb42 - fix : wrong job was retrieved + coverage was enabled in optionnal C++ builds

Compare with previous version

mentioned in commit fcb939c3

merged