small size network 中文意思是什麼
small size network
解釋
小型網路-
This article puts forward a solution named divide - assemble by deducing the size of bp neural network to overcome entering the local best point, the dividing process is that a big bp neural network is divided into several small bp neural networks, every small bp neural network can study alone, after all small bp neural networks finish their study, we can assemble all these small bp neural networks into the quondam big bp neural networks ; on the basis of divide - assemble solution, this article discusses the preprocessing of input species and how to deduce the size of bp neural network further to make it easy to overcome entering the local best point ; for the study of every small bp neural network, this article adopts a solution named gdr - ga algorithm, which includes two algorithms. gdr ? a algorithm makes the merits of the two algorithms makeup each other to increase searching speed. finally, this article discusses the processing of atm band - width distribution dynamically
本文從bp網的結構出發，以減小bp神經網路的規模為手段來克服陷入局部極小點，提出了bp神經網路的拆分組裝方法，即將一個大的bp網有機地拆分為幾個小的子bp網，每個子網的權值單獨訓練，訓練好以後，再將每個子網的單元和權值有機地組裝成原先的bp網，從理論和實驗上證明了該方法在解決局部極小值這一問題時是有效的；在拆分組裝方法基礎上，本文詳細闡述了輸入樣本的預處理過程，更進一步地減小了bp網路的規模，使子網的學習更加容易了；對于子網的學習，本文採用了最速梯度? ?遺傳混合演算法（即gdr ? ? ga演算法） ，使gdr演算法和ga演算法的優點互為補充，提高了收斂速度；最後本文闡述了用以上方法進行atm帶寬動態分配的過程。 -
Is an icann accredited internet top level domain name registrar located in china, focuses on providing network solution for small & medium size enterprises
是一個icann認證的網際網路頂級域名注冊商，在中國市場的主營業務是提供中小企業上網解決方案。 -
What we do at this aspect are : firstly, we describe the permutation symmetry of the structure of some special networks and the corresponding attractor sets with some geometric graphs in euclidean space, which are called attractors graph and geometrized structure graph of the networks respectively ; the geometrizing conditions are also given ; we study the dynamical behavior of the networks using the geometrized structure graph and attractors graph of the network ; moreover, we propose an approach to construct a big - size network with some small - size network with symmetry by the method of direct - sum, direct - produce and semidirect - produce. we also study the dynamical properties " relation between the big - size network and the small - size networks. all those results will provide some theoretical basis for designing a special large - scale network
本文在這方面所做的工作如下：首次將一些特殊網路的結構和吸引子集的置換對稱性用三維歐氏空間中的一些幾何圖來表示，分別稱之為幾何結構圖和吸引子圖；給出了網路對稱性的幾何化條州即相應的對稱性群為可遷群） ：並惜助網路的幾何結構圖和吸弓吁圖分析網路的動力學性質；此外，我們提出了用簡單的具有一定對稱性的小網路按照群的直和、半直積和直積的方式組合成較大的網路的方法，探討了這些小網路和所組成的大網路的一些動力學性質的關系，如穩定態的個數、各穩定態的回憶性質等，為較大網路的設計提供一些理論依據。 -
Being different from traditional neural network or nn, nn is based on traditional statistics, which provides conclusion only for the situation where sample size is tending to infinity, while svm is based on statistical learning theory or slt, which is a small - sample statistics and concerns mainly the statistic principles when sample are limited, especially the properties of learning procedure
支持向量機（ svm ）是九十年代中期發展起來的新的機器學習技術，與傳統的神經網路（ nn ）技術不同， svm是以統計學習理論（ slt ）為基礎， nn是以傳統統計學理論為基礎。傳統統計學的前提條件是要有足夠多的樣本，而統計學習理論是著重研究小樣本條件下的統計規律和學習方法的，它為機器學習問題建立了一個很好的理論框架。 -
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。 -
In this thesis, the analysis find four main problemes in the technology innovations in middle or small - size ecorporations. the first problem is the lackof funds, the second one is the lack of qulified scientists and technicians ; the third one is lack of network for medium and cooperation ; the last problem is that policy framework has not been formed, which support the technology innovation in middle or small - size ecorporations
本文分析了我國現階段中小企業在技術創新活動中存在的主要問題，一是資金來源不足嚴重製約了中小企業的技術創新；二是我國中小企業技術創新人才缺乏；三是缺乏有利於中小企業技術創新的中介和協作網路；四是政府對中小企業技術創新重視不夠，尚未形成扶持中小企業技術創新的政策框架。 -
With the drive of fap frames, the proposed method can realize realistic facial animation. the generated data is of small size, compatible with all mpeg - 4 system, also suitable for network transportation. the system realized with this method has great advantage for user ’ s operation
這種方法能夠在連續fap幀的驅動下實現真實感較強的三維人臉動畫，該方法數據量小，適合於網路傳輸，而且具有數據通用性強，計算復雜度低，實時性強，操作簡單等諸多優點。 -
The slide mechanism mainly by pushing for and the hydraulic power system with the necessary supporting wireless sensor detection and computer control system and network components, equipment is small in size, light forces, carrying capacity, a high degree of automation, convenient operation flexibility, security and reliability, the slip lifting the weight of components, and standard features such as unlimited distance
連續頂推機器人主要由頂推器、液壓動力系統以及與之配套的傳感檢測和計算機無線和網路控制系統等組成，具有設備體積小、自重輕、承載能力大，自動化程度高，操作方便靈活，安全可靠性好，被滑移、頂升構件的重量、尺度和距離不受限制等特點。 -
Some resource - limited small devices cannot join jini network directly due to the inadequate processor power and small memory size. in order to implement the function of plug - and - play for such devices, the jini surrogate architecture technology can be used
對于那些資源有限的小型設備，由於受處理器能力、內存大小的限制，沒有足夠的能力運行java虛擬機以加入jini網路者，這些設備要利用jini實現即插即用，必須採用jini代理體系結構技術。 -
Classification, as an important theme in data mining, has been researched earlier in statistics, machine learning, neural network, expert systems and etc. but most algorithms are memory resident, typically assuming a small data size
分類是數據挖掘中的一項非常重要的任務，目前在商業上應用很多。分類器的構造技術有統計方法、機器學習方法、神經網路方法等。大部分演算法是內存駐留演算法，適用於小型數據集。 -
The ssma series products are as little small - scale connection device that a kind of whorl joined, it has the characteristics that the size is small, width, light, dependability are high frequently, suitable for being small - scale electronic equipment and network system of high accuracy of the request the weight, volume
Ssma系列產品是一種螺紋連接的微小型連接器，它具有尺寸小、頻帶寬、重量輕、可靠性高的特點，適用於對重量、體積有要求的高精度的小型電子設備及網路系統。 -
It use a psc to replace the switch, and implement the csma / cd in the network interface card ( nic ). when the throughput grows and transfer speed is raised, csma / cd will lower the network ' s performance, but think about such truths below, the half - duplex gigabit optic ethernet will have some market space. first, there still exist large amount of small - size local network nowadays ; but they require high transmit speed, such as studio, lab etc, especially multimedia, graph and image processing, virtual reality and advertising
但是，在目前情況下，出於以下幾個方面的考慮，使用csma cd方式的半雙工千兆位網路也有它的發展空間：第一、在目前的情況下，有大量的小型局域網存在；但是又要求高的傳輸速率，如工作室、實驗室等工作組級的網路，尤其是現在十分常見的多媒體應用、圖形圖像處理、虛擬現實與廣告渲染等方面。 -
This paper firstly discusses the characteristics and advantages of the information transmission through sms provided by gsm network theoretically and systemically, discusses the necessity and possibility of the application of the technology into the power network detection information system, demonstrates the specifications and standards of the wireless communication interfaces used in the systems ; then, it demonstrates the constitution of the power network information system based on gsm network and the process of the operation of the system, based on this, proposes the project of one small size, cost - effective, convenient and high reliability remote detection and control unit
本文首先從理論上系統地論述了通過gsm網路提供的簡訊息服務sms進行信息傳輸的特點，討論了將該技術應用於電力網檢測信息系統中的可能性，對系統中用到的無線通信介面規范及標準進行了說明。闡述了基於gsm網路的電力網信息系統的組成，以及其工作過程。提出了一個體積小、經濟，方便用戶使用，可靠性高的遠端檢測控制單元實現方案。
分享友人