mirror of https://github.com/llvm/torch-mlir
6f548fc3ad
This commit adds the decomposition of `aten.adaptive_avg_pool2d` op into `aten.avg_pool2d` op. The current decomposition only supports cases where input size is equal to the output size. Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com> |
||
---|---|---|
.. | ||
basic.mlir | ||
elementwise.mlir | ||
flatten.mlir | ||
pooling.mlir | ||
unsqueeze.mlir |