An efficient and rapid Agrobacterium tumefaciens-mediated transformation protocol was developed to generate activation-tagged mutant lines with the aim of large-scale functional analysis of the potato genome. The expl...An efficient and rapid Agrobacterium tumefaciens-mediated transformation protocol was developed to generate activation-tagged mutant lines with the aim of large-scale functional analysis of the potato genome. The explants were inoculated with an Agrobacterium strain harboring the binary plasmid pSKI074 containing four CaMV 35S enhancers in the T-DNA region which activates the downstream genes in the host plant after its integration. Various parameters investigated to increase transformation efficiency were the type and age of explant, cultivar, hormone combinations, preculture of explants, period of co-cultivation with bacteria and concentration of bacterial cultures used for transformation. Stem explants from 5 week old plantlets of cv. Bintje which had undergone phytohormone pretreatment for 4 days, inoculation with diluted bacterial concentration of OD600 = 0.2 containing acetosyringone followed by 2 days of co-cultivation and selection in media with IAA and trans-zeatin all helped in greatly improving the transformation efficiency. The total time required from infection to rooted shoots was 6-7 weeks. Initial evidence for stable integration and expression of the transgenes by PCR analysis showed that over 93% of the regenerated lines were transgenic and this was confirmed by Southern hybridization.展开更多
为解决风电机组故障诊断中故障样本不足而导致模型准确率不高的问题,将当下备受关注的数据增强方法-去噪扩散概率模型(denoising diffusion probability model,DDPM)引入到故障诊断领域以生成大量高质量的故障样本数据集。因此,结合Tran...为解决风电机组故障诊断中故障样本不足而导致模型准确率不高的问题,将当下备受关注的数据增强方法-去噪扩散概率模型(denoising diffusion probability model,DDPM)引入到故障诊断领域以生成大量高质量的故障样本数据集。因此,结合Transformer网络,提出了一种DDPM-Transformer风电机组故障样本生成方法。首先,将用于计算机视觉图像生成领域的DDPM模型应用于风电机组故障诊断领域中,通过前向加噪过程将数据逐渐转化为噪声,再通过逆向去噪过程将噪声逐步恢复为原始数据,实现从噪声中生成故障数据,解决数据不平衡问题;其次,通过对原始DDPM中使用的U-net模块进行改进,使用Transformer模型替换U-net网络,利用扩散后的数据和添加的噪声训练Transformer模型,实现噪声预测,以提高故障数据的生成质量;最后,使用多种生成模型评价指标对生成的故障数据进行评价,在监督控制和数据采集系统(supervisory control and data acquisition,SCADA)故障数据生成中论证改进DDPM-Transformer模型的性能。通过试验证明,所提DDPM-Transformer模型与现有的生成模型相比,最大均值异(maximum mean discrepancy,MMD)最大提升0.13,峰值信噪比(peak signal to noise ratio,PSNR)最大提升7.8。所提模型可以有效地生成质量更高的风电机组故障样本,从而基于该样本集辅助训练基于深度学习的故障诊断模型,可以使诊断模型具有更高精度和良好的稳定性。展开更多
深度学习是人工智能领域的热门研究方向之一,它通过构建多层人工神经网络模仿人脑对数据的处理机制。大型语言模型(large language model,LLM)基于深度学习的架构,在无需编程指令的情况下,能通过分析大量数据以获得理解和生成人类语言...深度学习是人工智能领域的热门研究方向之一,它通过构建多层人工神经网络模仿人脑对数据的处理机制。大型语言模型(large language model,LLM)基于深度学习的架构,在无需编程指令的情况下,能通过分析大量数据以获得理解和生成人类语言的能力,被广泛应用于自然语言处理、计算机视觉、智慧医疗、智慧交通等诸多领域。文章总结了LLM在医疗领域的应用,涵盖了LLM针对医疗任务的基本训练流程、特殊策略以及在具体医疗场景中的应用。同时,进一步讨论了LLM在应用中面临的挑战,包括决策过程缺乏透明度、输出准确性以及隐私、伦理问题等,随后列举了相应的改进策略。最后,文章展望了LLM在医疗领域的未来发展趋势,及其对人类健康事业发展的潜在影响。展开更多
文摘An efficient and rapid Agrobacterium tumefaciens-mediated transformation protocol was developed to generate activation-tagged mutant lines with the aim of large-scale functional analysis of the potato genome. The explants were inoculated with an Agrobacterium strain harboring the binary plasmid pSKI074 containing four CaMV 35S enhancers in the T-DNA region which activates the downstream genes in the host plant after its integration. Various parameters investigated to increase transformation efficiency were the type and age of explant, cultivar, hormone combinations, preculture of explants, period of co-cultivation with bacteria and concentration of bacterial cultures used for transformation. Stem explants from 5 week old plantlets of cv. Bintje which had undergone phytohormone pretreatment for 4 days, inoculation with diluted bacterial concentration of OD600 = 0.2 containing acetosyringone followed by 2 days of co-cultivation and selection in media with IAA and trans-zeatin all helped in greatly improving the transformation efficiency. The total time required from infection to rooted shoots was 6-7 weeks. Initial evidence for stable integration and expression of the transgenes by PCR analysis showed that over 93% of the regenerated lines were transgenic and this was confirmed by Southern hybridization.
文摘为解决风电机组故障诊断中故障样本不足而导致模型准确率不高的问题,将当下备受关注的数据增强方法-去噪扩散概率模型(denoising diffusion probability model,DDPM)引入到故障诊断领域以生成大量高质量的故障样本数据集。因此,结合Transformer网络,提出了一种DDPM-Transformer风电机组故障样本生成方法。首先,将用于计算机视觉图像生成领域的DDPM模型应用于风电机组故障诊断领域中,通过前向加噪过程将数据逐渐转化为噪声,再通过逆向去噪过程将噪声逐步恢复为原始数据,实现从噪声中生成故障数据,解决数据不平衡问题;其次,通过对原始DDPM中使用的U-net模块进行改进,使用Transformer模型替换U-net网络,利用扩散后的数据和添加的噪声训练Transformer模型,实现噪声预测,以提高故障数据的生成质量;最后,使用多种生成模型评价指标对生成的故障数据进行评价,在监督控制和数据采集系统(supervisory control and data acquisition,SCADA)故障数据生成中论证改进DDPM-Transformer模型的性能。通过试验证明,所提DDPM-Transformer模型与现有的生成模型相比,最大均值异(maximum mean discrepancy,MMD)最大提升0.13,峰值信噪比(peak signal to noise ratio,PSNR)最大提升7.8。所提模型可以有效地生成质量更高的风电机组故障样本,从而基于该样本集辅助训练基于深度学习的故障诊断模型,可以使诊断模型具有更高精度和良好的稳定性。
文摘深度学习是人工智能领域的热门研究方向之一,它通过构建多层人工神经网络模仿人脑对数据的处理机制。大型语言模型(large language model,LLM)基于深度学习的架构,在无需编程指令的情况下,能通过分析大量数据以获得理解和生成人类语言的能力,被广泛应用于自然语言处理、计算机视觉、智慧医疗、智慧交通等诸多领域。文章总结了LLM在医疗领域的应用,涵盖了LLM针对医疗任务的基本训练流程、特殊策略以及在具体医疗场景中的应用。同时,进一步讨论了LLM在应用中面临的挑战,包括决策过程缺乏透明度、输出准确性以及隐私、伦理问题等,随后列举了相应的改进策略。最后,文章展望了LLM在医疗领域的未来发展趋势,及其对人类健康事业发展的潜在影响。
文摘随着算力网络中计算资源与虚拟化设备的广泛应用,在算力网络虚拟化中,针对云集群弹性伸缩策略基于阈值的响应式触发过程中存在的弹性滞后问题,提出一种基于Transformer的预测式云集群资源弹性伸缩方法(Predictive Cloud Cluster Resource Elastic Scaling Method Based on Transformer,Cloudformer).该方法利用序列分解模块将云集群数据分解为趋势项和季节项,趋势项采用双系数网络分别对输入空间预测的均值和方差进行归一化和反归一化,季节项采用融合傅里叶变换的频域自注意力模型进行预测,并在模型训练过程中使用指数移动平均模型动态调整训练损失的误差范围.实验结果表明,对比最先进的五种预测式弹性伸缩算法,本文所提出的方法在保持较低的模型训练和推理时间下,不同预测窗口单变量与多变量预测均方误差分别降低了10.07%和10.01%.
