Pix2Pix Tekki Nataal ci Nataal

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

1. Ban xeetu done tàggat la Pix2Pix soxla?
   • misaali nataal yuñ boole dugal-génne
   • Amul benn done
   • Bind yi kese
   • Enregistrement audio
2. Ban arsitektir generatër la Pix2Pix di jëfandikoo?
   • Reseau bu leer te boole lépp
   • U-Net bu am lëkkaloo skip
   • Reseau buy baaxoo
   • Garab buy jël dogal
3. Luy jàngat PatchGAN ci Pix2Pix?
   • Nataal bi yépp nekk benn poñ global
   • Dayo fichier nataal bi
   • Realism bu patch nataal yu ndaw
   • Jamonoy tàggat yaram
4. Lu tax Pix2Pix yokk L1 perte ci wetu perte adversarial?
   • ngir yeexal tàggat yaram
   • Ngir dindi kiy xeex
   • Defar bruit bu bari
   • To anchor pixelwise and low-frequency fidelity while the adversarial loss encourages sharper texture
5. Lan moo waral lëkkaloo U-Net yi di am njariñ ci tekki làkk yi?
   • Dañuy jaar ci jumtukaay buñ bokk ni ay boor ci biir reso bi
   • Dañuy efaase ay layer yu amul njariñ
   • Dañuy soppi nataal yi def ko mbind
   • Dañuy wàññi dayo done yi
6. Ban model moo yokk Pix2Pix ngir jàng ci dajale nataal yuñ boolewul?
   • Resnet
   • SikloGAN
   • Bert
   • Word2Vec
7. Why can a PatchGAN discriminator encourage sharper local image texture?
   • It predicts the input labels directly
   • It replaces the generator’s U-Net
   • It judges realism in small image patches rather than only scoring the whole image
   • It computes only global color histograms
8. In Pix2Pix, which part of the objective most directly anchors an output to its paired target image?
   • The learning-rate warmup
   • The GAN’s random seed
   • The PatchGAN receptive-field count
   • The L1 pixel-difference loss