Kediska Kordhinta Waqtiga Imtixaanka

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

1. Waqtiga go'aaminta, muxuu kordhinta-waqtiga-tijaabadu ku sameeyaa hal-gelin?
   • Ku socodsiinta moodalka go'an ee dhowr nuqul oo la beddelay oo wax gelinta ah iyo isku-darka wax-soo-saarka
   • Dib u tababarida qaabka xogta la kordhiyay mar kasta oo la waydiiyo
   • Ku darista buuqa miisaanka moodeelka ka hor saadaasha kasta
   • Cadaadiska moodeelka si uu dhakhso ugu socdo marka la eego
2. When can pooling predictions from transformed views reduce prediction variance?
   • When every view changes the true class
   • When the transforms preserve the label and the view errors are at least partly uncorrelated
   • When all predictions are identical by design
   • When the model is retrained on each transformed copy
3. Waa maxay qiimaha ugu weyn ee isticmaalka TTA?
   • Waxay u baahan tahay xogta tijaabada sumadeysan
   • Waxay si joogto ah u beddeshaa miisaanka la tababaray
   • Waxay kordhisaa xisaabinta fikradda sababtoo ah moodelku wuxuu socdaa dhawr jeer muunadkiiba
   • Waxay yaraynaysaa saxnaanta moodeelka agabyada nadiifka ah
4. Waa kuwee doorashada kordhinta ee dhab ahaantii dhaawici karta kala soocida nambarka?
   • Iftiin yar ayaa isbeddela
   • Dalagyo yar yar oo random
   • Ku darida wax yar oo Gaussian ah
   • Si toos ah u rogaaya '6' si ay ugu ekaato '9'
5. Qaybinta, waa maxay tillaabada dheeraadka ah ee loo baahan yahay ka hor inta aan la isku celcelin wax soo saarka TTA?
   • U roga isbeddel kasta oo joomatari ah si waji-xidhaha la saadaaliyay ay ula jaanqaadaan pixels-kii asalka ahaa
   • Dib-u-tababarka furaha ee muuqaal kasta
   • Tuurista dhammaan aragtiyaha la rogay
   • U beddelashada maaskaro sanduuqyada xidhitaanka
6. For a classification model, what may be aggregated across test-time views?
   • The model’s training labels
   • The predicted probabilities or logits from the views
   • The input image file names
   • The optimizer momentum buffers
7. Does standard test-time augmentation require retraining the model on new labels?
   • Yes; each view needs a newly trained model
   • Yes; it requires extra class labels for every view
   • No; it reuses fixed weights and evaluates transformed inputs at inference
   • No; it replaces the model with a lookup table
8. A segmentation model predicts masks on flipped and cropped views. What must happen before pixelwise pooling?
   • Average masks before undoing the transforms
   • Discard the original image coordinates
   • Apply a different label to each pixel view
   • Invert the geometric transforms so all masks align with the original image coordinates