statistics n. 1.統(tǒng)計(jì)學(xué),統(tǒng)計(jì)法〔用作單數(shù)〕。 2.統(tǒng)計(jì)數(shù)字[資料],統(tǒng)計(jì)表〔用作復(fù)數(shù)〕。 Government statisticsindicate that prices have gone down. 政府統(tǒng)計(jì)指出物價(jià)已經(jīng)下降。 You may consult the statistics on population issued by the government. 你可以查一查政府發(fā)表的人口統(tǒng)計(jì)。 the vital statistics (出生、結(jié)婚、死亡等)人口動(dòng)態(tài)統(tǒng)計(jì)。 collect [take] statistics 進(jìn)行統(tǒng)計(jì)。
sample n. 1.樣品,貨樣。 2.標(biāo)本;榜樣,實(shí)例。 3.【統(tǒng)計(jì)】典型取樣,抽檢查。 4.【電訊】信號(hào)瞬時(shí)值。 5.【冶金】 〔 pl. 〕 鋅華。 That is a fair sample of his manners. 那就是他的典型態(tài)度。 a light sample 光脈沖。 up to sample 1. 和樣品一樣[相符]。 2. 可以接受的,可以同意的。 vt. 從…取樣,從…抽樣;提供;貨樣;對(duì)…進(jìn)行抽樣檢查。
Acquisition of the fuse alarm digital variables : during idle , repeat acquired the 120 routes 48 analog data , process sample statistic for each rote acquired active - high 熔絲報(bào)警開關(guān)量信息的采集:在idle時(shí)候循環(huán)采集120路開入數(shù)據(jù),對(duì)各路采到的實(shí)時(shí)高電平次數(shù)進(jìn)行樣本統(tǒng)計(jì)。
After explaining the both at hone and abroad relevant results of study , the whole paper . taking the approaches of sample statistics , comparison study , and many other approaches of prediction , has carried on the deep discussion on how to establish the relationship between quantity aid quality in the development of higher education , how to establish the scientific view of higher education in quality , how to construct the rational structure of higher education , and how to raise the efficiency in school management of higher educatin and realize that the scale , quality , structure , effectiveness of higher education develop harmoniously and allsidedly 本文就湖北省高等教育大眾化的規(guī)模、結(jié)構(gòu)、質(zhì)量和效益問題進(jìn)行了全面系統(tǒng)的分析。全文采取了抽樣統(tǒng)計(jì)法、比較研究法、以及多種預(yù)測(cè)的研究方法,對(duì)國(guó)內(nèi)外有關(guān)的研究結(jié)果進(jìn)行闡述后,對(duì)如何確立高等教育發(fā)展中數(shù)量與質(zhì)量的關(guān)系、如何確立科學(xué)的高等教育質(zhì)量觀、如何構(gòu)建合理的高等教育結(jié)構(gòu)、以及如何提高高等教育辦學(xué)效益,實(shí)現(xiàn)高等教育規(guī)模、質(zhì)量、結(jié)構(gòu)、效益全面協(xié)調(diào)發(fā)展等問題,進(jìn)行了深入的探討。
In this paper , taking one of the special processes - chromic acids anodizing for example , such administration methods as the key points control and the quality performance administration system for surface treatment process have been setup by process analysis and key points seeking . as for the complicacy of the quality control on metallurgical special processes and the huge statistic data , the quality information system of surface treatment have been established on the bases of computer database . this paper brings out the systematic evaluation way to setup reception test items by taking the reception test results as evaluation basis and sampling statistics as rules 本文以鉻酸陽(yáng)極氧化表面處理工藝為研究對(duì)象,通過工藝技術(shù)分析、尋找關(guān)鍵特性、建立對(duì)關(guān)鍵特性的控制等工作,研究建立了表面處理工藝動(dòng)態(tài)質(zhì)量管理系統(tǒng);針對(duì)冶金特種工藝質(zhì)量控制的復(fù)雜性及統(tǒng)計(jì)數(shù)據(jù)量大等特點(diǎn),建立了以計(jì)算機(jī)數(shù)據(jù)庫(kù)管理為基礎(chǔ)的表面處理信息及處理系統(tǒng);為解決冶金特種工藝原材料入廠復(fù)驗(yàn)任務(wù)重的困難,建立了以原材料入廠復(fù)驗(yàn)的試驗(yàn)結(jié)果為評(píng)價(jià)基礎(chǔ),以抽樣統(tǒng)計(jì)方法為依據(jù)的原材料入廠復(fù)驗(yàn)項(xiàng)目?jī)?yōu)選系統(tǒng)等。
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(jī)( svm )是九十年代中期發(fā)展起來的新的機(jī)器學(xué)習(xí)技術(shù),與傳統(tǒng)的神經(jīng)網(wǎng)絡(luò)( nn )技術(shù)不同, svm是以統(tǒng)計(jì)學(xué)習(xí)理論( slt )為基礎(chǔ), nn是以傳統(tǒng)統(tǒng)計(jì)學(xué)理論為基礎(chǔ)。傳統(tǒng)統(tǒng)計(jì)學(xué)的前提條件是要有足夠多的樣本,而統(tǒng)計(jì)學(xué)習(xí)理論是著重研究小樣本條件下的統(tǒng)計(jì)規(guī)律和學(xué)習(xí)方法的,它為機(jī)器學(xué)習(xí)問題建立了一個(gè)很好的理論框架。