Idanwo ti o sopọ mọ itọsọna · Alabọde Ipele

Agbelebu-Afọwọsi adanwo

Check how fold-based validation estimates model performance, prevents leakage, and separates hyperparameter tuning from final evaluation.

Jẹmọ itọsọna awọn ọnaCross afọwọsi
Ibeere 1 ti 8

Ni boṣewa k-agbo agbelebu-afọwọsi, igba melo ni aaye data kọọkan lo fun idanwo?

Awotẹlẹ gbogbo ibeere8 ibeere

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

  1. Ni boṣewa k-agbo agbelebu-afọwọsi, igba melo ni aaye data kọọkan lo fun idanwo?
    • Gangan ni ẹẹkan
    • k igba
    • k-1 igba
    • O da lori awọn irugbin laileto
  2. Kini anfani akọkọ ti k-agbo agbelebu-afọwọsi lori ọkọ oju-irin kan / pipin idanwo?
    • O nigbagbogbo mu ki awoṣe išedede
    • O funni ni iduroṣinṣin diẹ sii, idiyele ti o gbẹkẹle orire ti gbogbogbo
    • O yọ iwulo fun eto idanwo kan kuro patapata
    • O ṣe iyara ikẹkọ awoṣe
  3. Stratified k-fold cross-afọwọsi jẹ iwulo paapaa nigbati:
    • Atọka data jẹ iwọntunwọnsi pipe
    • Awọn data ni a akoko jara
    • O ni awọn kilasi ti ko ni iwọntunwọnsi ati pe o fẹ ki agbo kọọkan tọju awọn iwọn kilasi ti o jọra
    • O fẹ lati ṣe ikẹkọ lori gbogbo dataset ni ẹẹkan
  4. Ijẹrisi-agbelebu-ọkan-jade (LOOCV) ṣeto nọmba awọn agbo-iwọn k dogba si:
    • 2
    • 10
    • Nọmba awọn ẹya ara ẹrọ
    • Nọmba awọn ayẹwo ni dataset
  5. Lati yago fun jijo data lakoko afọwọsi-agbelebu, awọn igbesẹ iṣaaju bi igbelowọn ẹya yẹ ki o jẹ:
    • Darapọ nikan inu agbo ikẹkọ kọọkan, lẹhinna lo si agbo afọwọsi
    • Dada lori ipilẹ data ni kikun ṣaaju pipin si awọn agbo
    • Fo patapata
    • Ti ṣe nikan lori awọn agbo afọwọsi
  6. Kini idi ti k-agbo agbelebu-ifọwọsi ko yẹ fun asọtẹlẹ jara-akoko?
    • Awọn jara akoko ni awọn ori ila diẹ ju
    • Awọn agbo airotẹlẹ le fi data iwaju sinu eto ikẹkọ ti a lo lati ṣe asọtẹlẹ ohun ti o kọja
    • O yara ju fun jara akoko
    • Time jara ko le wa ni pin ni gbogbo
  7. What is the purpose of nested cross-validation?
    • 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. Which fold counts are common starting choices for k-fold cross-validation?
    • Exactly two folds in every problem
    • One fold regardless of sample size
    • One thousand folds for every dataset
    • Five or ten folds