ONE - On-device Neural Engine
Loading...
Searching...
No Matches
OneHot.cc
Go to the documentation of this file.
1/*
2 * Copyright (c) 2025 Samsung Electronics Co., Ltd. All Rights Reserved
3 *
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
7 *
8 * http://www.apache.org/licenses/LICENSE-2.0
9 *
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 * See the License for the specific language governing permissions and
14 * limitations under the License.
15 */
16
17#include "../KernelGenerator.h"
18#include "../Validator.h"
19
20#include <AclKernelGen.h>
21
23{
24
25void Validator::visit(const ir::operation::OneHot &) { _supported = true; }
26
27void KernelGenerator::visit(const ir::operation::OneHot &node)
28{
29 const auto out_idx{node.getOutputs().at(0)};
30 const auto indices_idx{node.getInputs().at(ir::operation::OneHot::Input::INDICES)};
31 const auto depth_idx{node.getInputs().at(ir::operation::OneHot::Input::DEPTH)};
32 const auto onvalue_idx{node.getInputs().at(ir::operation::OneHot::Input::ON_VALUE)};
33 const auto offvalue_idx{node.getInputs().at(ir::operation::OneHot::Input::OFF_VALUE)};
34
35 auto output_tensor = _tensor_reg->getAclTensor(out_idx);
36 auto indices_tensor = _tensor_reg->getAclTensor(indices_idx);
37 auto depth_tensor = _tensor_reg->getAclTensor(depth_idx);
38 auto onvalue_tensor = _tensor_reg->getAclTensor(onvalue_idx);
39 auto offvalue_tensor = _tensor_reg->getAclTensor(offvalue_idx);
40
41 const size_t output_rank = _ctx.at(out_idx).shape().rank();
42 int32_t axis = node.param().axis == -1 ? output_rank - 1 : node.param().axis;
43 axis = acl_common::ToARMComputeAxis(output_rank, axis).value();
44
45 auto fn = acl_common::generateLayer<arm_compute::NEOneHot>(
46 indices_tensor->handle(), depth_tensor->handle(), onvalue_tensor->handle(),
47 offvalue_tensor->handle(), output_tensor->handle(), axis);
49}
50
51} // namespace onert::backend::acl_neon
std::unique_ptr< exec::IFunction > _return_fn
const Object & at(const Index &index) const
Get the object that is associated with the given index.
ARMComputeAxis ToARMComputeAxis(uint32_t rank, uint32_t axis)
Definition Swizzle.h:45
std::unique_ptr< AclFunction > asAclFunction(std::unique_ptr<::arm_compute::IFunction > &&layer)
Definition Convert.cc:246