基于Depth-wise卷积和视觉Transformer的图像分类模型

图像分类作为一种常见的视觉识别任务,有着广阔的应用场景。在处理图像分类问题时,传统的方法通常使用卷积神经网络,然而,卷积网络的感受野有限,难以建模图像的全局关系表示,导致分类精度低,难以处理复杂多样的图像数据。为了对全局关系进行建模,一些研究者将Transformer应用于图像分类任务,但为了满足Transformer的序列化和并行化要求,需要将图像分割成大小相等、互不重叠的图像块,破坏了相邻图像数据块之间的局部信息。此外,由于Transformer具有较少的先验知识,模型往往需要在大规模数据集上进行预训练,因此计算复杂度较高。为了同时建模图像相邻块之间的局部信息并充分利用图像的全局信息,提出了一种基于Depth-wise卷积的视觉Transformer(Efficient Pyramid Vision Transformer,EPVT)模型。EPVT模型可以实现以较低的计算成本提取相邻图像块之间的局部和全局信息。EPVT模型主要包含3个关键组件:局部感知模块(Local Perceptron Module,LPM)、空间信息融合模块(Spatial Information Fusion,SIF)和“+卷积前馈神经网络(Convolution Feed-forward Network,CFFN)。LPM模块用于捕获图像的局部相关性;SIF模块用于融合相邻图像块之间的局部信息,并利用不同图像块之间的远距离依赖关系,提升模型的特征表达能力,使模型学习到输出特征在不同维度下的语义信息;CFFN模块用于编码位置信息和重塑张量。在图像分类数据集ImageNet-1K上,所提模型优于现有的同等规模的视觉Transformer分类模型,取得了82.6%的分类准确度,证明了该模型在大规模数据集上具有竞争力。

Creative Problem-Solving and Music: Analyzing the Correlation between Music and Divergent Thinking Abilities

This research paper delves into the connection, between problem-solving and music. It’s a topic that has piqued the interest of scholars in fields, including science and neuroscience. The study explores how music can influence our ability to think divergently which is an aspect of creative thinking. It builds upon advancements in methods to investigate the relationship between music and divergent thinking aiming to uncover potential correlations. Doing it offers insights into the interplay between artistic expression and cognitive innovation. This research combines an analysis of existing literature with data collected from a group of participants shedding light on how music impacts our capacity for creative thinking. It demonstrates that music plays a role as a catalyst, for stimulating and enhancing thinking abilities.

初中科学数字化教学资源WISE“绘图故事”的研究

科学课程应重视图教学,还应积极开发相关的数字化教学资源。基于网络的科学探究环境WISE提供的“绘图故事”项目,采用知识整合理论设计探究过程,促进学生主动建构新知识;创设科学故事线,激发学生在真实探究情境中学习;依据学习目标设计数字资源序列,确保技术目标与探究学习目标相一致;针对学习阻障设计数字化认知工具,破解常规探究教学的难点;借助虚拟动画搭建表达和反思的支架,帮助学生自主探究。研究发现,采用知识整合理论设计数字化教学资源,可以有效促进八年级学生对运动图的理解并提高他们的知识整合水平,增益效果超过中等,为有效开发探究式初中科学数字化教学资源提供了具体指南,也为科学课程的图教学提供了数字化教学资源的样例。

Study on the Relationship Between College Students’ Self-Leadership and Career Preparation from the Perspective of Problem-Solving Skills

Based on the measurement of problem-solving skills,self-leadership,and career preparation,this paper collects 251 questionnaires from college students through online questionnaire survey.Through the analysis of the questionnaire results,this paper aims to clarify the relationship between college students’self-leadership and career preparation from the perspective of problem-solving skills,in order to provide ideas for college students’future career path and development.The results showed that from the perspective of problem-solving skills,there is a positive correlation between self-leadership and career preparation,and self-leadership has a positive impact on career preparation.

Be Wise with Money

Money plays an important role in our lives.Many of us receive pocket money from our parents,but how do we manage it?Do we spend it on everything we like,or do we use it wisely?Being wise with money is crucial,and here are some tips to help you do just that.

回顾ResourceWise对2023年纸浆和造纸工业的预测

随着2023年的结束,是时候反思一下ResourceWise全球销售副总裁Matt Elhardt对2023年造纸业的深刻预测了。让我们借用此文章重温一下这些预测,看看它们在过去一年中是如何展开的。1欧洲的能源价格波动将产生赢家和输家能源价格对纸浆和造纸的生产成本有重大影响。

Problem-Solving Based Intervention for Informal Caregivers: A Scoping Review

Background: Caregiving for someone is a huge task for informal caregivers. The caregiving role is often challenging and overwhelming for them. Stress, burden, and depression have been frequently identified as the negative caregiving experiences. This systematic review synthesized the available evidence for the problem-based intervention provided for caregivers as there is little insight about the effects of problem-based intervention on caregivers. Objectives: To describe: 1) the design of problem-solving intervention;2) the effects of problem-solving intervention for caregiver outcomes. Methods: This review followed Arksey and O’Malley’s methodological framework for conducting scoping reviews which entail setting research questions, selecting relevant studies, charting the data and synthesizing the results. FPRISMA guidelines were adopted and several databases were searched including MEDLINE;EMBASE;web of science. This review contains literature from 2012 to 2019 on problem-solving-based intervention which intended to caregivers. Results: 41 publications representing 27 unique problem-based interventions. Problem-solving-based intervention has different extent effects on caregiver emotion status, burden and quality of life. Conclusions: Problem-solving intervention is promising in caregiver intervention which could reduce caregiver depression, anxiety and burden.

Impact of self-directed learning readiness and learning attitude on problem-solving ability among Chinese undergraduate nursing students

Objective: To explore the effects of self-directed learning readiness and learning attitude on problem-solving ability among Chinese undergraduate nursing students. Methods: A convenience sampling of 460 undergraduate nursing students was surveyed in Tianjin, China. Students who participated in the study completed a questionnaire that included social demographic questionnaire, Self-directed Learning Readiness Scale, Attitude to Learning Scale, and Social Problem-Solving Inventory. Pearson’s correlation analysis was performed to test the correlations among problem-solving ability, self-directed learning readiness, and learning attitude. Hierarchical linear regression analyses were performed to explore the mediating role of learning attitude. Results: The results showed that learning attitude (r=0.338, P<0.01) and self-directed learning readiness (r=0.493, P<0.01) were positively correlated with problem-solving ability. Learning attitude played a partial intermediary role between self-directed learning readiness and problem-solving ability (F=74.227, P<0.01). Conclusions: It is concluded that nursing educators should pay attention on students’ individual differences and take proper actions to inspire students’ self-directed learning readiness and learning attitude.

Method and Apparatus for Creating Problem-Solving Complexes from Individual Elements

Based on the biological key-lock-principle common in various biological systems such as the human brain, this paper relates to a method and device for creating problem-solving complexes from individual elements that can be coupled with one another and that have different properties to solve problems. The problem solution can be carried out either serially with a large computer, or with several independent, hierarchically joined computers. In this system, an independent control unit that assumes a multitude of tasks and also acts as an interface with access to all participating computers, is assigned to each problem or object class according to the amount of potential problem-oriented solutions. Such a unit prepares the partial solutions found in its computer for the totality of the solutions computed in the associated computers, finally leading to a total problem solution.

Role of academic self-efficacy in the relationship between self-directed learning readiness and problem-solving ability among nursing students

Objective: Problem-solving should be a fundamental component of nursing education because It is a core ability for professional nurses. For more effective learning, nursing students must understand the relationship between self-directed learning readiness and problem-solving ability. The aim of this study was to investigate the relationships among self-directed learning readiness, problemsolving ability, and academic self-efficacy among undergraduate nursing students.Methods: From November to December 2016, research was conducted among 500 nursing undergraduate students in Tianjin, China,using a self-directed learning readiness scale, an academic self-efficacy scale, a questionnaire related to problem-solving, and selfdesigned demographics. The response rate was 85.8%.Results: For Chinese nursing students, self-directed learning readiness and academic self-efficacy reached a medium-to-high level,while problem-solving abilities were at a low level. There were significant positive correlations among the students' self-directed learning readiness, academic self-efficacy, and problem-solving ability. Furthermore, academic self-efficacy demonstrated a mediating effect on the relationship between the students' self-directed learning readiness and problem-solving ability.Conclusions: To enhance students' problem-solving ability, nursing educators should pay more attention to the positive impact of self-directed learning readiness and self-efficacy in nursing students' education.

Effects of Sex Differences and Problem-Solving Strategies on Digital Game Performance

Problem-solving strategy i,,; a critical skill in inquiry-based learning. Several studies have investigated how to use learning games to improve inquiry abilities. However, not every learner favors this kind of approach. Thus, we need to examine how human factors affect learners＇ reactions to the use of a digital game to support inquiry-based learning. This study addressed this issue by developing a digital game, ＂Baking Town＂, and using the game to examine the effects of two central human factors, sex differences and problem-solving strategies, on students＇ performance. The results demonstrate that students＇ inquiry abilities were significantly improved after they participated in the digital game. The results also demonstrate that the digital game may be a feasible way to reduce differences between boys and girls in this domain. Finally, we used the findings to develop a framework that can be used to enhance our understanding of sex differences and the use of problem-solving strategies in the context of digital games.

Circumstance Leading to Pregnancy and Problem-Solving Process of Unwanted Teenage Pregnancy in the Phetchaburi Province, Thailand

This descriptive qualitative study aims to understand circumstance leading to pregnancy and problem-solving process of unwanted teenage pregnancy in the Phetchaburi Province. A qualitative research design was used. Participants were 15 unwanted pregnant teenagers, aged between 15 and 19 years. All participants were primiparous and interviewed at postpartum stage. A semi-structured in-depth interview was the main technique of data collection. Data was analyzed using the process of manifest content analysis. Five themes, nine categories, 19 subcategories emerged from the data analysis. The themes were： （1） circumstance leading to pregnancy was the failure of contraceptive use and lack of knowledge about reproductive health; （2） negative emotional reactions included worry about parents＇ rejection and being blamed by others; （3） boyfriend＇s reaction to the pregnancy had a big impact on teen girl＇s feelings toward the situation. The majority of the teenagers＇ pregnancies accepted a child in the womb. The couples did not consider termination of the pregnancy; （4） parent＇s seeking a solution to serve the family dignity by organizing wedding ceremony and collaboratively plans for the future; and （5） life changes after the pregnancy due to the strong concerns about being blamed by the society. Findings from this study suggest that teens should particiPate in sex education classes that are specific for teenagers to prevent becoming pregnant. Moreover, the educational program specific for the teenagers＇ needs should be established so that they can continue their pregnancy without quitting school.

利用WISE平台对线上教学环境下探究式教学的实践探索

2022年4月21日《义务教育物理课程标准(2022年版)》(以下简称“新课标”)发布,本文以最新课标为指导,考虑线上线下结合式教学现状,面对线上教学背景下进行以学生为主体的探究实验时实验器材难以寻找、安全性难以保障、不能及时评价反馈等不可忽视的问题,借助WISE平台探索线上环境下一对多、多对多交互的探究活动进行方式,以线上教学环境下的小组合作探究活动为对象设计教学流程并实施,试图找到一种在线上教学环境下有效促使学生像科学家一样经历问题、证据、解释、交流等环节进行科学探索,促进学生物理学科核心素养的发展.

Enhancing the Problem-Solving Skills of Vocational Students Through Skills Competition

Unlike general education,vocational education aims to nurture skilled people for various sectors in the society.The development of students’problem analysis and problem-solving skills is crucial.As an important part of the vocational education system,skills competitions are considered a“booster”for improving the quality of teaching in vocational institutions.This paper examines the problem-solving skills of students in preparation for skills competition.First of all,we introduce China’s education policies and conduct a review of early scholars’views,followed by a discussion on the specific problems faced in the design of the entry for skills competition and an exploration of the process of enhancing students’problem-solving skills around these problems;we then propose several suggestions for vocational institutions to enhance their participation in skills competitions.Skills competitions provide a“special stage”for students in vocational institutions to show their abilities.The question of how this“stage”can be utilized to better improve students’abilities is worth exploring in different fields.

基于WISE的Webshell检测技术研究

本文主要在现有语义分析的基础上,基于机器学习技术,通过引入虚拟执行,构建多级检测机制,开发出一种具备更强大、更智能的检测能力的WISE智能语义分析引擎。

WISE:海尔创客赋能新模式

创客所有制体现共创、共治、共享、共赢的理念,让每个员工都有成就感。在中国企业转型升级过程中,鲜有通过赋能创客影响企业家认知的战略变革成功案例和理论研究,海尔创客赋能WISE模型提供了有益探索。创客赋能WISE模型WISE模型,即方向(Where).

“云”之动力,极“智”可靠——浅析科华数据Wise Power云动力预制式电力模组解决方案

近年来,随着数据中心承载的算力持续“狂飙”,其占国家总用电比例也大幅攀升,甚至被戏称为“不冒烟的钢厂”,而如何提高这座“数字钢厂”的含“绿”量,以更加低碳的基础设施来降低数据中心PUE,也成为业界关注的热点问题。科华数据股份有限公司(以下简称“科华数据”)立足电力电子产业高地,匠心研发Wise Power云动力预制式电力模组解决方案(以下简称“预制式电力模组”),致力于打造锚定数据中心“低碳靶心”的一枚利箭。

基于YOLOv5s−FSW模型的选煤厂煤矸检测研究

针对现有煤矸检测模型存在的特征提取不充分、参数量大、检测精度低且实时性差等问题,提出了一种基于YOLOv5s−FSW模型的选煤厂煤矸检测方法。该模型在YOLOv5s的基础上进行改进,首先将主干网络的C3模块替换为FasterNet Block结构,通过降低模型的参数量和计算量提高检测速度;然后,在颈部网络引入无参型SimAM注意力机制,增强模型对复杂环境下重要目标的关注,进一步提高模型的特征提取能力;最后,在输出端用Wise−IoU替换CIoU边界框损失函数,使模型聚焦普通质量锚框,提高收敛速度和边框的检测精度。消融实验结果表明:与YOLOv5s模型相比,YOLOv5s−FSW模型的平均精度均值(mAP)提高了1.9%,模型权重减少了0.6 MiB,参数量减少了4.7%,检测速度提高了19.3%。对比实验结果表明:YOLOv5s−FSW模型的mAP达95.8%,较YOLOv5s−CBC,YOLOv5s−ASA,YOLOv5s−SDE模型分别提高了1.1%,1.5%和1.2%,较YOLOv5m,YOLOv6s模型分别提高了0.3%,0.6%;检测速度达36.4帧/s,较YOLOv5s−CBC,YOLOv5s−ASA模型分别提高了28.2%和20.5%,较YOLOv5m,YOLOv6s,YOLOv7模型分别提高了16.3%,15.2%,45.0%。热力图可视化实验结果表明:YOLOv5s−FSW模型对煤矸目标特征区域更加敏感且关注度更高。检测实验结果表明:在环境昏暗、图像模糊、目标相互遮挡的复杂场景下,YOLOv5s−FSW模型对煤矸目标检测的置信度得分高于YOLOv5s模型,且有效避免了误检和漏检现象的发生。

基于改进YOLOv8n的变电设备红外图像实例分割算法

变电设备是电网输变电过程的重要组成部分,为保证电网的正常运行,需对变电设备进行故障诊断,红外图像中变电设备的精确分割是故障诊断的关键步骤。针对红外图像复杂场景中变电设备分割时存在的分割精度低和漏分割的问题,提出一种基于改进YOLOv8n的变电设备红外图像实例分割算法。首先设计一种上下文引导的特征增强下采样块替换YOLOv8n中的下采样卷积层,充分利用上下文信息和全局信息,增强模型对复杂场景的理解能力;然后引入可变形卷积重构Backbone中的C2f模块,增强对不规则设备特征的提取能力;最后用Wise-IOUv2对损失函数进行优化,提高模型的泛化性和分类能力。使用变电设备红外图像数据集对该模型进行实验验证,实验结果表明,相较于YOLOv8n基准模型,本文所提方法的mAP50和mAP50:95分别提高了4.2%和3.5%,所提方法能够较好地解决复杂场景下设备漏分割的问题,有效提高变电设备实例分割的准确率。

基于信息共享策略的加权逐点预测动态多目标优化

为了有效实现Pareto最优解,并且提升对于输入误差的鲁棒性,提出一种基于信息共享策略的加权逐点预测动态多目标优化算法。引入一种信息共享策略,该策略允许每个点利用其相邻解决方案中的信息有效提升模型的鲁棒性。引入一个相似性度量,并对其与一些常用的相似性度量进行对比分析,进一步提出一种加权逐点预测方法,逐点特性极大增强了捕捉各种模式的能力。在EC2018 DMO测试套件上的实验结果验证了该方法的有效性。