本文研究了使用可分析的学习样本来训练神经网络的可行性问题。
This paper investigates the feasibility of using analytically generated training samples to train neural networks.
以实际工程21组数据为学习样本,对工程中5个点进行了计算。
The neural network model studied 21 samples and calculate 5 locations.
提出了一种基于统计的学习样本生成方法,使样本生成问题规范化。
An improved BP algorithm and a learning sample generation method based on statistics are performed.
实验表明,该方法对500个未学习样本的识别率达到了95.8%。
Experiments demonstrate that the method achieves a recognition rate of 95.8 % in 500 unstudied samples.
在无监督学习中,只向网络提供一些学习样本,而不提供理想的输出。
In unsupervised learning, only learning to network with some samples, rather than provide an ideal output.
然而对于学习样本较多、输入输出映射关系复杂的情况,学习速度较慢。
However, its learning speed is slow in learning more advanced copies and a complicated output-input projection relationship.
摘要:规则学习算法通过学习样本产生规则集,如伺判断规则集的好坏?
Absrtact: the rule extraction algorithm produces the rule set by learning examples. How to evaluate the rule set?
通过训练神经网络学习样本数据,建立了正确的发动机怠速神经网络模型。
By using the collection data in training the neural network, this paper establishes the proper engine idle speed neural network model.
讨论了模型的学习样本、网络参数对预测精度的影响,选出最佳网络参数配置。
The affection of learning samples and network parameters on prediction accuracy was discussed, the best network parameters were selected.
在大量的学习样本基础上,结果表明该模型能够快速、准确、可靠地预测材料性能。
A large number of samples in the study, based on the model results show that the fast, accurate and reliable prediction of material properties.
每个粒子的学习样本包括全局最优粒子、自身最优粒子和粒子邻居中最优运行粒子;
Then the learning mechanism of each particle was separated into three parts: its own historical best position, the best neighbor and the global best one.
以生产数据为学习样本,建立BP神经网络模型,实现对剪切旋压工艺参数的预测。
BP neutral networks model has been established in order to predict process parameter of shear spinning based on many data accumulated during manufacturing.
对18组地下矿山的特征参数值作为学习样本进行训练和检验,回判估计的误判率为0。
Taking 18 sets of parameter data of underground mine as training and testing samples, the ratio of mis-discrimination is 0.
其次,提出了基于属性效用函数估计的学习样本构造方法,从决策问题本身抽取学习样本。
Secondly, to extract learning samples from the MADM problem, an approach to estimate the utility functions for attributes is presented.
为了克服历史数据不足的问题,设计了通过时间序列聚类分析进行学习样本集的积累的方法。
To overcome the shortage of historical data, the increment of learning samples are got by clustering analysis the time series data from Ticket sale record.
实验表明,在小学习样本条件下SVM比RBF人工神经网络具有更好的分类性能和推广能力。
The experiment shows that SVM processes better classification performance and spreading potential than RBF manual neural network under the small study sample condition.
实验结果表明,该方法检测效果好,且要求学习样本少,适用于不同缺陷类型和各种检测问题。
Experiments demonstrated that this approach has good detection ability performance and needs less learning samples, which makes it suitable for many types of defect and textured material.
由设计出的基于BP神经网络的智能自整定PID控制器的神经网络部分对学习样本进行学习。
The part of neural network of the intelligent self-tuning PID controller based on BP network learns the learning sample.
可以看到在较好的选取学习样本情况下,神经网络技术在电子透镜的逆设计中有着明显的优越性。
In the condition of selecting the learning samples properly, the artificial neural network has the obvious advantage in the inverse designing the electronic lens.
支持向量机方法基于小学习样本条件下,通过寻求结构风险最小,以期获得良好的分类效果和泛化能力。
The SVM method is based on seeking on the Structural Risk Minimization by few learning samples supporting, and it has important feature such as good generalization and classification performance, etc.
将某城市1997年已处理结案的典型交通事故案例作为学习样本,构建了智能化的交通事故责任认定模型。
By using a number of closed cases in certain Cities in 1997 as learning object, the intellectualized traffic accident liability judgment model is proposed. As a new deci...
将某城市1997年已处理结案的典型交通事故案例作为学习样本,构建了智能化的交通事故责任认定模型。
By using a number of closed cases in certain Cities in 1997 as learning object, the intellectualized traffic accident liability judgment model is proposed. As a new decision support...
SVM分类器的学习样本是模仿各种多矢量曲线变换和攻击下,相应产生了由多曲线检测相关值构成的特征向量。
The study samples of SVM classification are imitating all kinds of eigenvector, which are detected from the attacked and transformed curves.
把1975到2006年全国的客运量数据和其他相关指标作为学习样本,验证寻优参数得到训练模型预测结果的可靠性。
Taking the 1975-2006 National passenger traffic data and other related indicators as a learning sample, then verify the validity of the training model after Parameter Optimization.
以14个边坡工程的稳定状况作为学习样本和预测样本,讨论了基于神经网络技术的黄土边坡稳定性分析方法及其有效性。
Regarding stability conditions of 14 slopes as study samples and predicting samples, we discuss the stability analysis method and its usefulness based on neural network technology.
依据《大学生体育合格标准》,建立起神经网络的学习样本,提出了基于神经网络的的男子大学生身体素质评估的训练模型。
University student physical quality test evaluation model based on neural network was set up (according) to "University student Sports Qualification Standard".
采用统一、规范化准则,对各优化目标进行等级划分并进行不同的组合,形成FNN的学习样本集,其输出即决策解的综合性能的评估值。
According to the unified and standardization criterion, the optimal objectives are graded and combined to form learning sample set for FNN model, and its output is the integrated evaluated value.
采用人工神经网络的BP算法,以电火花微小孔加工工艺参数正交实验的结果作为神经网络的学习样本,建立电火花微小孔加工多目标工艺参数的预测模型。
Though choosing the experimental results as the learning sample, the performance predictive model of EDM micro-and-small holes is proposed, with the BP algorithm of artificial neural network.
采用人工神经网络的BP算法,以电火花微小孔加工工艺参数正交实验的结果作为神经网络的学习样本,建立电火花微小孔加工多目标工艺参数的预测模型。
Though choosing the experimental results as the learning sample, the performance predictive model of EDM micro-and-small holes is proposed, with the BP algorithm of artificial neural network.
应用推荐