Cross-Validation
Nkwenye n'ofe bụ usoro ntugharịgharị maka ịtụpụta etu ihe nlereanya ga-esi gbasaa na data a na-ahụghị anya.
Nchịkọta
It makes better use of limited data and gives a more reliable performance estimate than a single train/test split.
Ime miri emi
Otu mgbawa ụgbọ oloko/ule na-esighi ike: akara ị nwetara na-adabere kpamkpam na ahịrị ndị dabara na ntọala ule. Cross-validation na-edozi nke a site n'ịtụgharị ọrụ nke setịpụrụ ule. Na k-fold cross-validation, ị na-ekewa data ahụ n'ime k nha nhata folds, na-azụ na k-1 n'ime ha, nyochaa n'ogige ejidere, ma na-emegharị k ugboro ka a na-anwale ahịrị ọ bụla otu ugboro. Nkezi akara k na-ewepụta atụmatụ kwụsiri ike yana ngbanwe nke mgbanwe. Nhọrọ ndị a na-ahụkarị bụ mpịakọta 5 ma ọ bụ 10. Ndị dị iche iche gụnyere k-fold stratified (ichekwa oke klaasị maka data enweghị aha), ịhapụ otu-apụ (k ha nhata ọnụọgụ nlele), yana usoro oge na-agbawa nke na-adịghị azụta ọdịnihu maka ịkọ oge gara aga.
Nghọta nka nka
Cross-validation kasị ike maka ihe nlereanya nhọrọ na hyperparameter n'iji ya gee ntị: ị na-atụnyere nhazi site na nkezi nkwado akara ha kama ịfefe na otu nkewa. Ọnyà dị oke egwu bụ ntapu data - nhazi ọ bụla nke 'na-ahụ' setịpụ data niile (nchịkọta, nhọrọ njirimara, nkọwa) ga-adabara n'ime mpịakọta ọ bụla, ọ bụghị tupu kewaa, ma ọ bụ atụmatụ gị ga-abụ nke enweghị nchekwube. Nkwenye obe na-ekewapụ nlegharị anya na nyocha ikpeazụ ka ịzenarị ntapu a.
Mmetụta atụmatụ
Mkpebi doro anya
Ọ na-enyere gị aka ikewapụta nkwupụta ọrụ aka doro anya na asụsụ ahịa.
Ọnụ ego na mmefu ego
Ị nwere ike ịjụ ajụjụ mmejuputa iwu ka mma tupu itinye ego ma ọ bụ oge.
Team na usoro ọrụ
Ndị otu nwere nghọta na-eme ka ngwaahịa, amụma na mkpebi mmụta ka mma.
Ọdịnihu nke Cross-Validation
Ka datasets na ụdị na-eto eto, ịgba ọsọ k n'usoro ọzụzụ zuru oke na-adị ọnụ, yabụ ndị na-eme ya na-enwewanye mmasị na otu nnukwu nkwado edobere maka mmụta miri emi ebe na-edobe nkwado ndabere maka obere dataset ma ọ bụ tabular. ML akpaghị aka na ngwaọrụ dị ka scikit-learn's GridSearchCV na Optuna na-eme nkwado ndabere n'ime ọchụchọ hyperparameter na ndabara. Nnyocha na-aga n'ihu na nsonye dị ọnụ ala karịa, pipeline na-eguzogide ọgwụ, yana nkwado kwesịrị ekwesị maka mkpokọta, usoro nhazi na data dabere na oge.
Mmejuputa n'ezie n'ụwa
Iji 5-fold cross-validation tulee mgbagha mgbagha, oke ọhịa na-enweghị usoro, na nkwalite gradient tupu ịmebe otu ụdị.
Itinye k-fold gbatịrị agbatị na dataset nchọpụta aghụghọ na-ezighi ezi ka akụkụ nke ọ bụla na-edobe otu oke klaasị na-adịghị ahụkebe.
Na-agba ọsọ GridSearchCV ma ọ bụ RandomizedSearchCV, nke na-akwado njikọ hyperparameter ọ bụla iji họrọ ntọala kacha mma.
N'iji usoro oge (mgbagharị / n'ihu-chaining) nkwenye cross-validation iji nyochaa ngwaahịa ma ọ bụ onye na-ebu amụma na-enweghị ọzụzụ na data n'ọdịnihu.
Ihe ize ndụ & okporo ụzọ nche
Otu dị iche iche nwere ike iji otu okwu ahụ mee ihe n'ụzọ dị iche, yabụ kọwapụta oge n'oge.
Ihe nrịbama nwere ike ịdị ike ebe arụmọrụ ụwa na-adaghị adaba.
Ileghara ogo data na atụmatụ nyocha anya na-emepụtakarị nsonaazụ na-adịghị mma.
Map mmejuputa
Malite na nkọwa asụsụ dị larịị nke nsonaazụ ịchọrọ.
Họrọ otu metrik ịga nke ọma na otu ọnọdụ ọdịda tupu nnwale.
Gbaa obere onye na-anya ụgbọ elu nwere data nnọchite anya, ọ bụghị ihe ngosi ngosi na-egbu maramara.
Detuo ebe Cross-Validation na-enyere aka yana ebe ụzọ ndị dị mfe dị mma.
Nọgide na-eme nchọpụta
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the Cross-Validation quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
Ntuziaka na-esote
Cross-Encoders vs Bi-Encoders
Ajụjụ a na-ajụkarị
What is Cross-Validation?
Nkwenye n'ofe bụ usoro ntugharịgharị maka ịtụpụta etu ihe nlereanya ga-esi gbasaa na data a na-ahụghị anya. Ọ na-eji obere data eme ihe nke ọma ma na-enye atụmatụ arụmọrụ a pụrụ ịdabere na ya karịa otu mgbawa ụgbọ oloko/ule.
Na ọkọlọtọ k-fold cross-validation, ugboro ole ka ejiri ebe data ọ bụla maka nnwale?
Mpịakọta ọ bụla na-eje ozi dị ka ule setịpụrụ kpọmkwem otu ugboro n'ofe k, ya mere, a na-anwale nlele ọ bụla n'otu oge ma zụọ ya na ndị ọzọ.
Gịnị bụ isi uru nke k-fold cross-validation n'elu otu ụgbọ oloko/ule kewara?
Site na nkezi karịa ọtụtụ mpịaji, nkwenye gafere na-ebelata ọdịiche nke atụmatụ arụmọrụ ma e jiri ya tụnyere ịdabere na otu nkewa aka ike.
Kedu ihe kpatara nkwado k-fold nkịtị na-ekwesịghị ekwesị maka ịkọ amụma usoro oge?
Ịtụgharị data n'usoro oge na-eme ka ọdịnihu bụrụ ọzụzụ, ya mere CV usoro oge na-eji nkewa na-aga n'ihu na-azụ naanị na nleba anya gara aga.