quiz bu lëkkale ak tegtal · Diggante Niveau

Focal Loss ngir quiz biy gis lu jaarul yoon

Saytu modulatëru focal-loss, desekilibraasu klaas, cër alpha ak gamma, ak ni misaal yiñ klaase bu baax ak yu dëgër yi di wuutee.

Laaj 1 de 8

Ban jafe-jafe bu mag ci gis mbir ci benn etap lañu defar ngir saafara focal loss?

Xoolaatal laaj bu nekk8 laaj

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

  1. Ban jafe-jafe bu mag ci gis mbir ci benn etap lañu defar ngir saafara focal loss?
    • Désequilibre bu tar diggante mbir yu bari yi am ci kanam ak ginaaw bu yomb
    • Gaawaayu inferensi bu yeex ci detektër yu am ñaari etap
    • Ñàkum done yuñ etikete ci wàllu tàggat yaram
    • Overfitting ci mbir yu rëy
  2. Lan mooy facteur modulateur bi perte focal di yokk ci entropie croix?
    • log (p_t)
    • (1 - p_t)^gamma
    • p_t^2 kese
    • 1 / p_t
  3. Lan mooy xew ci ñàkka focal sudee parametru focalisation gamma bi dafa def 0?
    • Dafay nekk njuumte kaare moyen
    • Dafay baña bàyyi xel ci misaal yu baax yépp
    • Dafay nekk entropie croix
    • Dafay dem ba amul àpp
  4. Lan la yokk gamma di def ci misaal yu yomb te ñu xaaj bu baax?
    • Yokk seen xaalis ci ñàkk
    • Duñu soppi seen wàll
    • Dindi leen ci done yi
    • Dafay tere seen contribution perte gëna am doole
  5. Ban arsitektir detektër moo dugal ñàkka focal?
    • RetinNet
    • YOLOv1
    • R-CNN bu gëna gaaw
    • AlexNet
  6. Ban cër la paramet alpha di faral di def ci wetu gamma?
    • Dafay tëral ni ñuy jàngee
    • Dafay ekilibre diisaay bi ci diggante klaas yu baax ak yu baaxul
    • Dafay saytu dayo nataal bi
    • Dafay màndargaal limu boyet ancrage yi
  7. For a hard example with p_t close to zero, what is the focal modulating factor (1 − p_t)^gamma likely to do?
    • Drive the factor to zero
    • Turn the example into a positive label
    • Keep the factor relatively close to one so the example retains substantial loss weight
    • Replace the cross-entropy term with alpha
  8. A dense detector has many easy background locations. What effect does focal loss aim to have on their combined training influence?
    • Increase their count by duplicating them
    • Remove all positive object examples
    • Make every example receive an identical loss
    • Reduce their dominance so training can focus more on sparse hard examples