mirror of https://github.com/alibaba/MNN.git
53 lines
1.6 KiB
C++
53 lines
1.6 KiB
C++
//
|
|
// ShapeOneHot.cpp
|
|
// MNN
|
|
//
|
|
// Created by MNN on 2019/11/29.
|
|
// Copyright © 2018, Alibaba Group Holding Limited
|
|
//
|
|
|
|
|
|
#include "shape/SizeComputer.hpp"
|
|
#include "core/Macro.h"
|
|
|
|
namespace MNN {
|
|
class ShapeOneHot : public SizeComputer {
|
|
public:
|
|
virtual bool onComputeSize(const MNN::Op* op, const std::vector<Tensor*>& inputs,
|
|
const std::vector<Tensor*>& outputs) const override {
|
|
MNN_ASSERT(4 == inputs.size());
|
|
auto indices = inputs[0];
|
|
auto depthTensor = inputs[1];
|
|
|
|
const int depth = depthTensor->host<int>()[0];
|
|
if (depth < 0) {
|
|
return false;
|
|
}
|
|
const int indicesDimension = indices->dimensions();
|
|
const int outputDimension = indicesDimension + 1;
|
|
|
|
auto param = op->main_as_OneHotParam();
|
|
int axis = param->axis();
|
|
if (axis < 0) {
|
|
axis = outputDimension + axis;
|
|
}
|
|
auto output = outputs[0];
|
|
output->buffer().dimensions = outputDimension;
|
|
output->buffer().type = inputs[2]->buffer().type;
|
|
for (int i = 0; i < outputDimension; ++i) {
|
|
if (i < axis) {
|
|
output->setLength(i, indices->length(i));
|
|
} else if (i == axis) {
|
|
output->setLength(i, depth);
|
|
} else {
|
|
output->setLength(i, indices->length(i - 1));
|
|
}
|
|
}
|
|
TensorUtils::getDescribe(output)->dimensionFormat = TensorUtils::getDescribe(inputs[0])->dimensionFormat;
|
|
return true;
|
|
}
|
|
};
|
|
|
|
REGISTER_SHAPE_INPUTS(ShapeOneHot, OpType_OneHot, (std::vector<int>{1}));
|
|
} // namespace MNN
|