mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
52 lines
1.8 KiB
C++
52 lines
1.8 KiB
C++
/*
|
|
* SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
|
|
* SPDX-License-Identifier: Apache-2.0
|
|
*
|
|
* 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 "tensorrt_llm/batch_manager/common.h"
|
|
#include "tensorrt_llm/runtime/bufferManager.h"
|
|
#include "tensorrt_llm/runtime/iTensor.h"
|
|
#include "tensorrt_llm/runtime/modelConfig.h"
|
|
#include "tensorrt_llm/runtime/promptTuningParams.h"
|
|
#include "tensorrt_llm/runtime/worldConfig.h"
|
|
|
|
namespace tensorrt_llm::batch_manager
|
|
{
|
|
|
|
class PromptTuningBuffers
|
|
{
|
|
|
|
public:
|
|
using SizeType32 = tensorrt_llm::runtime::SizeType32;
|
|
using ITensor = tensorrt_llm::runtime::ITensor;
|
|
using TensorPtr = runtime::ITensor::SharedPtr;
|
|
|
|
runtime::PromptTuningParams mPromptTuningParams;
|
|
SizeType32 mMaxPromptVocabSize;
|
|
|
|
PromptTuningBuffers(SizeType32 maxBatchSize, runtime::BufferManager const& manager,
|
|
runtime::ModelConfig const& modelConfig, runtime::WorldConfig const& worldConfig);
|
|
|
|
void validate(std::optional<TensorPtr> const& optReqPromptEmbeddingTable,
|
|
std::optional<SizeType32> const& optReqPromptVocabSize);
|
|
|
|
void fill(RequestVector const& contextRequests, RequestVector const& genRequests,
|
|
runtime::BufferManager const& manager, bool packed);
|
|
};
|
|
|
|
} // namespace tensorrt_llm::batch_manager
|