pytorch/caffe2/operators/numpy_tile_op.cc
James Reed 20cd27da42
[caffe2][ONNX] Implement CPU NumpyTileOp and corresponding ONNX backend (#7053)
* Implement CPU NumpyTileOp

* Address comments
2018-04-27 19:58:15 -07:00

19 lines
503 B
C++

#include "caffe2/operators/numpy_tile_op.h"
namespace caffe2 {
REGISTER_CPU_OPERATOR(NumpyTile, NumpyTileOp<CPUContext>);
OPERATOR_SCHEMA(NumpyTile)
.NumInputs(2)
.Input(0, "data", "The input tensor.")
.Input(1, "repeats", "1-D Tensor specifying how many times to repeat"
" each axis.")
.Output(
0,
"tiled_data",
"Tensor that will contain input replicated along the given axis.")
.InheritOnnxSchema("Tile");
} // namespace caffe2