diff options
author | Lei Zhang <antiagainst@google.com> | 2021-02-09 08:51:55 -0500 |
---|---|---|
committer | Lei Zhang <antiagainst@google.com> | 2021-02-09 08:55:20 -0500 |
commit | 068bf9e80220329f7baeaf8df34fb4818ad8aa58 (patch) | |
tree | dad07d2e511aef438184056ef17db0b66eea41d4 /clang/lib/Frontend/CompilerInvocation.cpp | |
parent | 4c640e49c9553363bc0e6fcbdbfe8d678683db97 (diff) | |
download | llvm-068bf9e80220329f7baeaf8df34fb4818ad8aa58.zip llvm-068bf9e80220329f7baeaf8df34fb4818ad8aa58.tar.gz llvm-068bf9e80220329f7baeaf8df34fb4818ad8aa58.tar.bz2 |
[mlir][linalg] Define a depthwise 2-D convolution op
This commit defines linalg.depthwise_conv_2d_nhwc for depthwise
2-D convolution with NHWC input/output data format.
This op right now only support channel multiplier == 1, which is
the most common case.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D94966
Diffstat (limited to 'clang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions