In the field of natural language processing(NLP),there have been various pre-training language models in recent years,with question answering systems gaining significant attention.However,as algorithms,data,and comput...In the field of natural language processing(NLP),there have been various pre-training language models in recent years,with question answering systems gaining significant attention.However,as algorithms,data,and computing power advance,the issue of increasingly larger models and a growing number of parameters has surfaced.Consequently,model training has become more costly and less efficient.To enhance the efficiency and accuracy of the training process while reducing themodel volume,this paper proposes a first-order pruningmodel PAL-BERT based on the ALBERT model according to the characteristics of question-answering(QA)system and language model.Firstly,a first-order network pruning method based on the ALBERT model is designed,and the PAL-BERT model is formed.Then,the parameter optimization strategy of the PAL-BERT model is formulated,and the Mish function was used as an activation function instead of ReLU to improve the performance.Finally,after comparison experiments with traditional deep learning models TextCNN and BiLSTM,it is confirmed that PALBERT is a pruning model compression method that can significantly reduce training time and optimize training efficiency.Compared with traditional models,PAL-BERT significantly improves the NLP task’s performance.展开更多
针对移动机器人领域自适应蒙特卡洛定位算法(Adaptive Monte Carlo Localization,AMCL)在相似及变化场景下易失效的问题,本文提出基于改进YOLOv8构建语义链表为AMCL提供预定位位姿的方法,改变粒子权重更新方式,进而提升定位准确性和鲁棒...针对移动机器人领域自适应蒙特卡洛定位算法(Adaptive Monte Carlo Localization,AMCL)在相似及变化场景下易失效的问题,本文提出基于改进YOLOv8构建语义链表为AMCL提供预定位位姿的方法,改变粒子权重更新方式,进而提升定位准确性和鲁棒性.以YOLOv8为基础,结合信息聚集-分发机制和注意力尺度序列融合模块增强其Neck部分特征融合能力,并对模型进行剪枝,提升精度和速度;利用激光SLAM(Simultaneous Localization And Map-ping)构建二维栅格地图,通过改进的YOLOv8提取物体语义并映射到地图上,得到二维语义地图,根据各连续语义物体之间的关系构建语义链表;在定位过程中,将机器人识别到的物体语义信息与语义链表进行匹配,为AMCL提供预定位位姿,改变其粒子更新方式进行精确定位,并基于词袋模型降低免疫障碍物遮挡导致的语义链断裂.在相似及变化场景下进行定位对比实验,实验结果验证了本文算法的有效性.展开更多
针对印刷电路板(Printed Circuit Board,PCB)表面缺陷检测任务中模型体积和参数量较大的问题,提出了一种基于通道剪枝的轻量级YOLOv8n网络PCB缺陷检测算法。为有效提升对PCB小目标缺陷的特征提取能力,采用RepViT作为特征提取网络;为提...针对印刷电路板(Printed Circuit Board,PCB)表面缺陷检测任务中模型体积和参数量较大的问题,提出了一种基于通道剪枝的轻量级YOLOv8n网络PCB缺陷检测算法。为有效提升对PCB小目标缺陷的特征提取能力,采用RepViT作为特征提取网络;为提升网络对小目标的关注度,减少神经网络推理过程中的梯度信息重复,将颈部网络的卷积模块替换为Rep-Net with Cross-Stage Partial CSP and ELAN(RepNCSPELAN4);为降低缺陷重叠时检测框失真现象,在预测部分使用Focaler-MPDIoU替换完全交并比(Complete Intersection over Union,CIoU);利用层自适应幅度分数剪枝(Layer Adaptive Magnitude based Pruning,LAMP)方法对融合改进方法的模型进行修剪,去除模型中冗余的梯度信息和权重,减少参数量和浮点运算量,压缩模型体积。实验结果表明,在PCB公开数据集中,经过LAMP之后,该算法相较于YOLOv8n,参数量下降60.8%,模型体积减小50.8%,计算量下降48.8%,平均精度均值(mean Average Precision,mAP)提高3.8%。在提高精度的同时,计算量、参数量和模型体积都低于原模型,满足在低配置设备下的使用需求。展开更多
针对当前番茄叶片病害检测模型参数量、计算量过大的问题,该研究提出了一种基于YOLOv8n的轻量化高精度网络模型。通过StarBlock模块对原始的C2f(CSP bottleneck with 2 convolutions)进行重构,大幅降低参数量的同时增强模型表达能力;其...针对当前番茄叶片病害检测模型参数量、计算量过大的问题,该研究提出了一种基于YOLOv8n的轻量化高精度网络模型。通过StarBlock模块对原始的C2f(CSP bottleneck with 2 convolutions)进行重构,大幅降低参数量的同时增强模型表达能力;其次引入混合局部通道注意力机制(mixed local channel attention,MLCA),以捕捉更多的上下文信息和多尺度特征;同时,通过多级通道压缩方式改进了原有检测头,减少了沿通道维度的特征;最后通过融合通道剪枝算法对模型二次压缩,使其更加轻量化。试验结果表明,经处理的模型参数量、浮点计算量、模型权重大小分别降低了63.3%、72.8%、61.9%,模型精确率、召回率和平均精度均值(mean average precision(IoU=0.5),mAP_(0.5))分别为97.5%、96.2%和98.5%,性能方面,移动端设备检测帧率达到358.5帧/s,番茄叶片病虫害图像单幅推理时间平均为4.4 ms。证明了该算法可在大幅降低网络计算量的同时保持较高的检测性能,能够满足移动端和嵌入式设备的部署要求。展开更多
基金Supported by Sichuan Science and Technology Program(2021YFQ0003,2023YFSY0026,2023YFH0004).
文摘In the field of natural language processing(NLP),there have been various pre-training language models in recent years,with question answering systems gaining significant attention.However,as algorithms,data,and computing power advance,the issue of increasingly larger models and a growing number of parameters has surfaced.Consequently,model training has become more costly and less efficient.To enhance the efficiency and accuracy of the training process while reducing themodel volume,this paper proposes a first-order pruningmodel PAL-BERT based on the ALBERT model according to the characteristics of question-answering(QA)system and language model.Firstly,a first-order network pruning method based on the ALBERT model is designed,and the PAL-BERT model is formed.Then,the parameter optimization strategy of the PAL-BERT model is formulated,and the Mish function was used as an activation function instead of ReLU to improve the performance.Finally,after comparison experiments with traditional deep learning models TextCNN and BiLSTM,it is confirmed that PALBERT is a pruning model compression method that can significantly reduce training time and optimize training efficiency.Compared with traditional models,PAL-BERT significantly improves the NLP task’s performance.
文摘针对移动机器人领域自适应蒙特卡洛定位算法(Adaptive Monte Carlo Localization,AMCL)在相似及变化场景下易失效的问题,本文提出基于改进YOLOv8构建语义链表为AMCL提供预定位位姿的方法,改变粒子权重更新方式,进而提升定位准确性和鲁棒性.以YOLOv8为基础,结合信息聚集-分发机制和注意力尺度序列融合模块增强其Neck部分特征融合能力,并对模型进行剪枝,提升精度和速度;利用激光SLAM(Simultaneous Localization And Map-ping)构建二维栅格地图,通过改进的YOLOv8提取物体语义并映射到地图上,得到二维语义地图,根据各连续语义物体之间的关系构建语义链表;在定位过程中,将机器人识别到的物体语义信息与语义链表进行匹配,为AMCL提供预定位位姿,改变其粒子更新方式进行精确定位,并基于词袋模型降低免疫障碍物遮挡导致的语义链断裂.在相似及变化场景下进行定位对比实验,实验结果验证了本文算法的有效性.
文摘针对印刷电路板(Printed Circuit Board,PCB)表面缺陷检测任务中模型体积和参数量较大的问题,提出了一种基于通道剪枝的轻量级YOLOv8n网络PCB缺陷检测算法。为有效提升对PCB小目标缺陷的特征提取能力,采用RepViT作为特征提取网络;为提升网络对小目标的关注度,减少神经网络推理过程中的梯度信息重复,将颈部网络的卷积模块替换为Rep-Net with Cross-Stage Partial CSP and ELAN(RepNCSPELAN4);为降低缺陷重叠时检测框失真现象,在预测部分使用Focaler-MPDIoU替换完全交并比(Complete Intersection over Union,CIoU);利用层自适应幅度分数剪枝(Layer Adaptive Magnitude based Pruning,LAMP)方法对融合改进方法的模型进行修剪,去除模型中冗余的梯度信息和权重,减少参数量和浮点运算量,压缩模型体积。实验结果表明,在PCB公开数据集中,经过LAMP之后,该算法相较于YOLOv8n,参数量下降60.8%,模型体积减小50.8%,计算量下降48.8%,平均精度均值(mean Average Precision,mAP)提高3.8%。在提高精度的同时,计算量、参数量和模型体积都低于原模型,满足在低配置设备下的使用需求。
文摘针对当前番茄叶片病害检测模型参数量、计算量过大的问题,该研究提出了一种基于YOLOv8n的轻量化高精度网络模型。通过StarBlock模块对原始的C2f(CSP bottleneck with 2 convolutions)进行重构,大幅降低参数量的同时增强模型表达能力;其次引入混合局部通道注意力机制(mixed local channel attention,MLCA),以捕捉更多的上下文信息和多尺度特征;同时,通过多级通道压缩方式改进了原有检测头,减少了沿通道维度的特征;最后通过融合通道剪枝算法对模型二次压缩,使其更加轻量化。试验结果表明,经处理的模型参数量、浮点计算量、模型权重大小分别降低了63.3%、72.8%、61.9%,模型精确率、召回率和平均精度均值(mean average precision(IoU=0.5),mAP_(0.5))分别为97.5%、96.2%和98.5%,性能方面,移动端设备检测帧率达到358.5帧/s,番茄叶片病虫害图像单幅推理时间平均为4.4 ms。证明了该算法可在大幅降低网络计算量的同时保持较高的检测性能,能够满足移动端和嵌入式设备的部署要求。
