TensorRT-LLMs/cpp/tensorrt_llm/runtime/generationConfig.h
2024-08-13 22:34:33 +08:00

59 lines
2.1 KiB
C++

/*
* Copyright (c) 2022-2024, NVIDIA CORPORATION. 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 "tensorrt_llm/runtime/common.h"
#include "tensorrt_llm/runtime/iTensor.h"
namespace tensorrt_llm::runtime
{
class GenerationConfig
{
public:
GenerationConfig() = default;
explicit GenerationConfig(SizeType32 batchSize, SizeType32 beamWidth, SizeType32 maxInputLength,
std::vector<SizeType32> maxAttentionWindowVec, SizeType32 maxAttentionWindow, SizeType32 sinkTokenLength,
SizeType32 maxSeqLength, SizeType32 inputLengthSum = SizeType32(0))
: batchSize{batchSize}
, beamWidth{beamWidth}
, maxInputLength{maxInputLength}
, maxAttentionWindowVec{maxAttentionWindowVec}
, maxAttentionWindow{maxAttentionWindow}
, sinkTokenLength{sinkTokenLength}
, maxSeqLength{maxSeqLength}
, inputLengthSum{inputLengthSum}
{
}
SizeType32 batchSize{};
SizeType32 beamWidth{};
SizeType32 maxInputLength{};
std::vector<SizeType32> maxAttentionWindowVec{};
SizeType32 maxAttentionWindow{};
SizeType32 sinkTokenLength{};
SizeType32 maxSeqLength{};
SizeType32 inputLengthSum{}; // Initialized only if inputPacked is set to true in fromInput.
static GenerationConfig fromInput(ITensor const& inputIds, ITensor& inputLengths, bool inputPacked,
SizeType32 beamWidth, std::vector<SizeType32> maxAttentionWindowVec, SizeType32 maxAttentionWindow,
SizeType32 sinkTokenLength, SizeType32 maxSequenceLength);
};
} // namespace tensorrt_llm::runtime