选择变量，建立回归模型，代入样本数据进行回归分析。
Changing variable, set up a regression model, take regression analyse using sample data.
这二者的改变最终都增加了采集样本数据所需的总时间。
Both of these changes result in an increase in the total time necessary to collect the sample of data.
注：本表数据为2006年人口变动情况抽样调查样本数据。
Note: figures in this table are sample figures in 2006 sample survey on population change.
迭代计算时，考虑样本数据值的统计特性、数据序列间距关系。
The method takes into account the statistics of the sample data and relationship among the annotated point series.
百分比抽样转换通过选择转换输入行的百分比来创建样本数据集。
The Percentage Sampling transformation creates a sample data set by selecting a percentage of the transformation input rows.
对参数特征值估计和可信区间的诠释都是得出样本数据推论的路径。
Point Estimates of parameters and Confidence Interval Interpretation are both means for making inferences about sample data.
基于这两个变量之间的关系，我们从我们的样本数据中得出的异常值。
Based on the relationship of these two variables, we derive outliers from our sample data.
不过，它们得到的数据确实与其中一次冬季直升机调查得到的样本数据相当一致。
However, the narwhal measurements do correlate well with one-shot samples taken by winter helicopter surveys.
然后通过教师样本数据对网络进行充分的训练，获得适宜的参数矩阵；
Second train the network well though training-samples and obtained the best network parameters vector.
第二、建立文本分类模型，使用大量的有害信息样本数据训练分类模型。
Secondly, Building text categorization model, and training the model by a great many harmful information samples data.
最后取一组小样本数据进行计算，实例结果表明所提出的方法合理有效。
At last an example is given to demonstrate the availability and ap-plication of the proposed method.
支持向量机基于结构风险最小化原则，解决了小样本数据分类和泛化问题。
SVM can solve small sample problems and has good generalization ability using the principles of structural risk minimization.
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.
由于本文只是一篇介绍性的文章，而且样本数据偏小，因此这里不打算就此话题展开讨论。
Because this is an introductory article and the sample data is small, the topic is not covered here.
该方法具有较强的稳健性，适应异常情况下的样本数据，能保持较满意预测精度。
This method has more strong robustness, suits the sample data in singular conditions and can keep a certain forecasting precision.
我们用从婆罗洲北部收集的63套尺蠖蛾样本数据，给这个想法一个严格的检验。
We provide a rigorous test of this idea, using a compilation of 63 samples of geometrid moths from northern Borneo.
结果表明，在训练集样本数据较少时，广义回归神经网络的预测准确度仍然很高。
The results showed that the prediction accuracy is satisfied, even though there are a few data in training sets.
世界上已开发了容有400件样本数据以上的模型数据库，并配置由沉积学家试验。
The prototype database with more than 400 sample data was developed and distributed for exandtiation by sedimentologists in the world.
引入减法聚类算法对样本数据进行分类，用得到的分类数据对局部模型参数进行离线辨识。
By introducing the subtraction clustering algorithm, the sample data are classified and the local model parameters are identified off-line using the corresponding data set.
预警模型对已知的样本数据可以完全判别，在初步使用后对未知样本数据的基本判断准确。
The early-warning model can distinguish the known sample data completely. After the preliminary use, the early-warning model can accurate distinguish almost the unknown sample data.
在样本数据范围内，当模型的收敛精度为0.1%时，识别和预测的泛化精度均在5%以内。
Within the sample database range, generalization precisions of identification and prediction are within 5% when convergence precision is within 0.1%.
有许多 推论统计步骤（根据样本数据估计总体参数的方法）可以被充分利用，但目前却没有应用它们。
An array of inferential statistical procedures (methodologies for estimating population parameters based upon sample data) could be fruitfully exploited, but at present are not being applied.
这就是为什么我那么热衷于comScore混合了固定样本数据和像素数据跟踪的综合方法。
And that is why I am so enthusiastic about comScore's hybrid approach of marrying panel data and tracking pixel data.
利用映射图象可以直观地剔除异常样本数据、选择优化决策变量和预测最优操作点和优化方向。
Using this mapping image can visually eliminate abnormal sample data, select decision variables and predict optimal operating point and direction for an industrial process.
马尔柯夫链适用于大样本数据序列的短期预测，而灰色系统预测方法适用于小样本数据的中期预测。
Markov Chain is suitable for short-term forecast of great capacity sample data sequence, but gray system forecast method is suitable for medium-term forecast of few capacity sample data sequence.
利用误差反向传播的改进算法对样本数据进行训练，并用另外的一些样本数据验证模型的应用效果。
Using the improved error backward propagation, the model is trained with stylebook data and validated its effect by other stylebook data.
通过高精度的数控移动工件台获取密集的样本数据，并在神经网络训练过程中采用贝叶斯正则化方法。
Dense sample data are acquired by using numerical control platform of high precision, and the Bayesian generalization is adopted during training the neural network.
改进了神经网络模型的算法，使其在样本数据相对较少的条件下也能很好地在性态上模拟被逼近函数。
The Neural Networks models is improved, it can simulate the approached function commendably even at the case of the little number of the samples.
改进了神经网络模型的算法，使其在样本数据相对较少的条件下也能很好地在性态上模拟被逼近函数。
The Neural Networks models is improved, it can simulate the approached function commendably even at the case of the little number of the samples.
应用推荐