wehub-resource-sync
34 lines
1.4 KiB
C++
34 lines
1.4 KiB
C++
// Copyright (c) 2025 PaddlePaddle Authors. All Rights Reserved.
|
|
//
|
|
// Licensed under the Apache License, Version 2.0 (the "License");
|
|
// you may not use this file except in compliance with the License.
|
|
// You may obtain a copy of the License at
|
|
//
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
//
|
|
// Unless required by applicable law or agreed to in writing, software
|
|
// distributed under the License is distributed on an "AS IS" BASIS,
|
|
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
// See the License for the specific language governing permissions and
|
|
// limitations under the License.
|
|
|
|
#pragma once
|
|
|
|
#include "paddle/phi/core/dense_tensor.h"
|
|
|
|
namespace phi {
|
|
template <typename T, typename Context>
|
|
void MoePermuteKernel(const Context &dev_ctx,
|
|
const DenseTensor &X,
|
|
const optional<DenseTensor> &XScale,
|
|
const DenseTensor &expert_routemap_topk,
|
|
const DenseTensor &expert_prob_topk,
|
|
const int num_experts,
|
|
const std::vector<int> &tokens_per_expert,
|
|
const int padding_multiplex,
|
|
DenseTensor *X_unzipped,
|
|
DenseTensor *zipped_expertwise_rowmap,
|
|
DenseTensor *token_prob_unzipped,
|
|
DenseTensor *XScale_unzipped);
|
|
} // namespace phi
|