training sample 中文意思是什麼
training sample
解釋
訓練樣本-
First, realized a wegener - willie distribute based network traffic anomaly detection algorithm. we make use of wegener - willie distribute to analyze the inherent time - frequency distribution characteristics of the traffic flow signal. then according to the experience of analysis on historical flow, we construct a normal flow training sample aggregation and a abnormal flow training sample aggregation
通過魏格納-威利分佈分析網路流量信號在時頻分佈上所反映出的內在特點，根據歷史流量的經驗構造正常流量和異常流量兩個訓練樣本空間，通過k最近鄰分類演算法將帶檢測流量信號的時頻分佈與訓練樣本進行比較，完成對檢測樣本的自動分類識別。 -
By using the air temperature measured by thermocouples and the glass temperature measured by infrared thermometer as the training sample, a model for prediction of the temperature parameters for the float glass in lehr was created based on the ameliorated bp neural network
摘要利用熱電偶測得的退火窯中空氣溫度和紅外測溫儀測得的玻璃表面溫度作為訓練樣本，建立了基於bp神經網路的玻璃溫度預測模型。 -
During re - estimating the e - hmm parameters, every training sample is represented by one e - hmm, the model parameters for every sample are obtained firstly, then the different model parameters were synthesized to one model through weighted method, and the weights are adaptively calculated in the training stage
重估模型參數時，首先計算每幅臉像相對應的模型參數，然後進行加權合併，權值由迭代公式求得，訓練結束後用一個合成的模型來表示一個對象。 -
In order to verify the feasibility of ann, adopting same training sample the author establishes quadratic curve model and index model of tourism foreign exchange income and cubic curve model and index model of total inbound tourist quantity
為了驗證人工神經網路模型的可行性，筆者用同樣的訓練樣本分別建立了旅遊外匯收入二次曲線模型、指數曲線模型和入境遊客三次曲線模型、指數曲線模型。 -
In chapter 5 the distributed cfar detection is studied when lds are correlated. due to the correlation of local decisions is not known, empirical estimation is adopt to resolve this problem, the relationship of the training sample size and the estimated confidence is analyzed
對局部採用硬判決的情況，深入研究了傳統的分析方法，利用經驗估計解決了局部觀測相關系數未知時存在的困難，分析了系統判決精度與估計樣本數量的關系。 -
The training sample, valid sample and test sample were developed, taking data of 26 sample plots from may to october of 2001 as input indicators and phytoplankton biomass of each sample site as output factor
以2001年5 ~ 10月全太湖26個采樣點的實測水文、水質、氣象等資料作為輸人因子，建立了訓練樣本、檢驗樣本和測試樣本，並以各采樣點的浮游植物量作為輸出因子。 -
Then, in the main system, neural network was adopted to construct the response relation between impeller performance and meridional channel design variables, where the training sample data were schemed according to the design of experimental method, and the effect of blade shape on the impeller performance was taken into account and the meridional channel was optimized
然後，在中心系統內利用試驗設計理論安排訓練樣本，採用神經網路建立葉輪性能與子午流道設計參數之間的響應關系，同時計及葉片形狀對葉輪性能的影響，對子午流道及葉片進行優化。 -
The formative training sample data are used to train the ann model
形成了預警系統的訓練樣本，對神經網路進行訓練。 -
Through frequent testing on the temperature of the objects, the training sample assemblies are obtained
通過對場景中物體的表觀溫度進行多次測量，得到訓練樣本集合。 -
In order to acquire the training sample, 12 impress modulus must be induced from 9 actual elastic parameters
要想獲得訓練樣本，就要由實際的九個彈性參數反推門個壓痕模量。 -
Classification is a sort of supervised learning ( i. e., the learning of the model is " supervised " in that it is told to which class each training sample belongs )
需要指出的是：分類是一種有指導的學習（即模型的學習在被告知每個訓練樣本屬于哪個類的「指導」下進行） 。 -
The numbers of the training sample of the networks are 37, 32, 27, 22 and the numbers of the test sample are 5, 10, 15, 20, respectively, and the relative errors of the predications are less than 3 %. it is shown that the rough network is accurate and available for prediction of stock market
訓練樣本集的天數分別是37天， 32天， 27天和22天，預測樣本集的天數分別是后5天， 10天， 15天和20天，預測的相對誤差都小於3 。 -
This paper summarized a system feature set for transient stability classification, several methods for analyzing the separability of input space of transient stability classification are discussed, tabu search - ing technique is employed to select an effective set of features from a large initial features set. the classification test shows that the presented method works very well for feature selection. fisher linear recognition is employed to cut down the training sample set, the computation burden of the ann training is alleviated very much, so the convergence performance is improved
本文總結提出了一組用於穩定分類的系統特徵，研究了幾種暫態穩定分類輸入空間可分性分析的方法，並利用tabu搜索技術從一個維數較大的特徵集中選擇出一組有效特徵，取得了良好的效果；研究提出了利用fisher線性識別技術壓縮訓練樣本集的方法，大大減輕了ann的訓練負擔，提高了ann收斂的性能。 -
After researching the application of artificial neural network to fault diagnosis of transformer, the back propagation algorithm is improved observably. inducting chaos dynamics principium to create chaos neural network back propagation algorithm and using the pretreatment of training sample, the emulation program show that this method is effective
本文深入研究了人工神經網路在變壓器故障診斷中的應用，對目前最常採用的bp演算法做出了較大改進，在原有演算法中引入混沌動力學原理構造了混沌神經網路bp演算法，並提出了訓練樣本集的預處理方法，在模擬訓練中取得了滿意的效果。 -
Normal behavior and anomaly are distinguished on the basis of observed datum such as network flows and audit records of host. when a training sample set is unlabelled and unbalanced, attack detection is treated as outlier detection or density estimation of samples and one - class svm of hypersphere can be utilized to solve it. when a training sample set is labelled and unbalanced so that the class with small size will reach a much high error rate of classification, a weighted svm algorithm, i
針對訓練樣本是未標定的不均衡數據集的情況，把攻擊檢測問題視為一個孤立點發現或樣本密度估計問題，採用了超球面上的one - classsvm演算法來處理這類問題;針對有標定的不均衡數據集對于數目較少的那類樣本分類錯誤率較高的情況，引入了加權svm演算法-雙v - svm演算法來進行異常檢測;進一步，基於1998darpa入侵檢測評估數據源，把兩分類svm演算法推廣至多分類svm演算法，並做了多分類svm演算法性能比較實驗。 -
Based on the principle of that whether the human behaviour is benefit to fire evacuation or not, seven input variables of training sample and one output variables were first normalized. fire reporting, fire identifying and fire extinguishing selected as the human behaviour output variables, this work has developed the 50 iterative error bar charts through 50 iterative training checkouts
根據調查對象的各種行為是否有利於逃生的原則，首次對訓練樣本的7個輸入變量和1個輸出變量做歸一化處理，選擇了報警求助、通知他人火災信息、嘗試救火三個有代表性的行為反應為輸出變量，經50代訓練與檢驗，並對行為反應變量用50代檢驗誤差建立直方圖。 -
This method utilizes an on - line algorithm based upon lssvm ( least square support vector machine ), which can build adaptive models to predict the cod values of unknown water samples quickly and accurately. in the modeling process, every training sample is also assigned a prior weight to take their significance to the final predictive model into account
該方法是一種基於最小二乘支持向量機的在線自適應加權演算法，這種演算法可以自適應地選取和未知水樣最相近的標準樣本進行建模，同時在建模中又利用加權的方法分別考慮了各個標準樣本重要性。 -
When prediction with little training sample set and large variable si ? is concerned, this paper abstracts the prime factors from the training sample set, then only inputs the prime factors into ann instead of primary variables
對于多指標小樣本預測問題，文中利用主成分分析法對原有指標體系進行處理，提取主成分構成新的指標作為神經網路的輸入。 -
Using error back propagation algorithm. the limitation of the bp net is also improved ; 2 > when prediction with little training sample set and large variable size is concemed, this paper abstracts the prime factors from the training sample set using rough set theory, then only inputs the prime factors into ann. this diminishes the size and the input nodes number of ann
對于即網路本身易陷入局部極小點及收斂速度慢的問題，在文中也得到了改進； 2 、對于財政轉移支付中標準收入的測算屬于多指標小樣本的預測問題，文中首次利用粗集理論對初始的指標體系進行約簡，提取出關鍵性的因素作為神經網路的輸入。 -
After every modular e - hmm is trained, the modular e - hmm parameters are combined into one e - hmm to represent one person. the advantage of the modular in training method is that the e - hmm parameters re - estimating process has good ability of adaptation : when new sample sets are added to the training sample, the information of the new sample sets can be conveniently combined into the e - hmm, and the computational work is reduced. besides, the modular training method provides an answer for the problem of choice initial e - hmm parameters
本方法的優點是使e - hmm參數估計過程具有良好的靈活性和易擴充性：對新增加的訓練樣本，產生的最後模型既保留了已經訓練過的信息又能反映新增樣本的信息，同時減少了計算量，而且可以把訓練樣本和模型參數分開，在模型庫中只保留訓練好的模型參數值和有關的中間值，不必保留訓練樣本。
分享友人