Maswali ya Kuongeza Muda wa Mtihani

Majibu na maelezo huonekana unapojibu maswali.

1. At inference time, what does test-time augmentation do with a single input?
   • Kuendesha muundo uliowekwa kwenye nakala kadhaa zilizobadilishwa za ingizo na kuchanganya matokeo
   • Kufunza tena kielelezo kwenye data iliyoongezwa kila wakati inapoulizwa
   • Kuongeza kelele kwa uzani wa mfano kabla ya kila utabiri
   • Kushinikiza mfano ili iendeshe haraka kwa uelekezaji
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. Je, gharama kuu ya kutumia TTA ni nini?
   • Inahitaji data ya majaribio yenye lebo
   • Inabadilisha kabisa uzani uliofunzwa
   • Inaongeza hesabu ya uelekezaji kwa sababu modeli huendesha mara nyingi kwa sampuli
   • Inapunguza usahihi wa mfano kwenye pembejeo safi
4. Ni chaguo gani la uongezaji linaweza KUUMIZA kiainishaji tarakimu?
   • Mwangaza mdogo hubadilika
   • Mazao madogo ya nasibu
   • Kuongeza kelele kidogo ya Gaussian
   • Kugeuza '6' kwa mlalo ili ionekane kama '9'
5. Katika sehemu, ni hatua gani ya ziada inahitajika kabla ya wastani wa matokeo ya TTA?
   • Inageuza kila mabadiliko ya kijiometri ili vinyago vilivyotabiriwa vijipange upya kwa saizi asili
   • Kufunza upya avkodare kwa kila mwonekano
   • Inatupa maoni yote yaliyogeuzwa
   • Kubadilisha barakoa kuwa masanduku ya kufunga
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