Kediska Xaqiijinta

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

1. Heerarka k- laab-laabashada ansaxinta, imisa jeer ayaa dhibco kasta oo xog ah loo isticmaalaa tijaabada?
   • Dhab ahaan hal mar
   • k jeer
   • k-1 jeer
   • Waxay kuxirantahay abuur random
2. Waa maxay faa'iidada ugu weyn ee k-laab isdhaafsiga ansaxinta hal tareen/kala qaybsanaan tijaabo ah?
   • Had iyo jeer waxay kordhisaa saxnaanta moodeelka
   • Waxay ku siinaysaa xasillooni badan, qiyaasta nasiib-ku-tiirsanaanta guud
   • Waxay meesha ka saaraysaa baahida loo qabo tijaabada gebi ahaanba
   • Waxay dedejisaa tababarka moodeelka
3. Xaqiijinta isdhaafsiga k-laab ee la habeeyey ayaa si gaar ah faa'iido u leh marka:
   • Xogta xogta ayaa si fiican u dheelli tiran
   • Xogtu waa wakhti taxane ah
   • Waxaad leedahay fasallo aan dheeli tirnayn oo waxaad doonaysaa in laalaab kastaa ilaaliyo saamiga fasalka la midka ah
   • Waxaad rabtaa in aad ku tababarto kaydka xogta oo dhan hal mar
4. Oggolaanshaha ansaxinta-hal-baxa (LOOCV) waxay dejisaa tirada laalaabyada k siman:
   • 2
   • 10
   • Tirada sifooyinka
   • Tirada shaybaarada ee ku jira xogta
5. Si looga fogaado daadinta xogta inta lagu guda jiro xaqiijinta isdhaafsiga, diyaarinta tillaabooyinka sida qaabaynta muuqaalku waa inay noqotaa:
   • Geli kaliya gudaha laalaab kasta oo tababar ah, ka dibna codso laalaabyada xaqiijinta
   • Ku habboon kaydka xogta oo dhan ka hor inta aanad u kala bixin laalaab
   • Gebi ahaanba waa la booday
   • Lagu sameeyay kaliya laalaabka ansixinta
6. Waa maxay sababta k-laabka caadiga ah ee ansaxinta iskutallaabtu aanay ku habboonayn saadaalinta taxanaha wakhtiga?
   • Taxanaha wakhtiga ayaa leh saf aad u yar
   • Laalaabyada aan sugnayn waxay gelin karaan xogta mustaqbalka shaxda tababarka ee loo isticmaalo in lagu saadaaliyo wixii la soo dhaafay
   • Aad bay ugu degdeg badan tahay taxanaha wakhtiga
   • Taxanaha wakhtiga gabi ahaanba lama kala qaybin karo
7. Waa maxay ujeedada ka gudubka ansixinta buulku?
   • It makes every model use the same hyperparameters
   • It lets preprocessing use the full dataset before splitting
   • It separates hyperparameter selection from an outer evaluation of the selected procedure
   • It removes the need for held-out evaluation
8. Waa kuwee laalaabyada tirinta ayaa ah xulashooyinka bilawga ah ee caadiga ah ee k-laab xaqiijinta iskutallaabta?
   • Exactly two folds in every problem
   • One fold regardless of sample size
   • One thousand folds for every dataset
   • Five or ten folds