Pix2Pix Mubvunzo-ku-Mufananidzo Wekushandura

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

1. Ndeupi rudzi rwe data rekudzidzisa rinodiwa nePix2Pix?
   • Mienzaniso yemifananidzo yakabatanidzwa
   • Hapana data zvachose
   • Zvinyorwa zvemashoko chete
   • Zvakarekodhwa
2. Ndeipi jenareta yekuvaka iyo Pix2Pix inoshandisa?
   • Yakajeka yakabatana zvizere network
   • U-Net ine skip yekubatanidza
   • A recurrent network
   • Muti wesarudzo
3. Musarura wePatchGAN muPix2Pix anotarisei?
   • Mufananidzo wacho wose sechibodzwa chimwe chepasi rose
   • Saizi yefaira remufananidzo
   • Chokwadi chezvigamba zvidiki zvepanzvimbo
   • Nguva yekudzidzira
4. Sei Pix2Pix ichiwedzera kurasikirwa kweL1 pamwe nekurasikirwa kweanopikisa?
   • Kudzikamisa kudzidzisa nemaune
   • Kubvisa kusarura
   • Kugadzira ruzha rwakangoitika
   • To anchor pixelwise and low-frequency fidelity while the adversarial loss encourages sharper texture
5. Sei maU-Net skip connections achinyanya kubatsira pamabasa ekushandura?
   • Ivo vanopfuudza yakagovaniswa chimiro semipendero yakananga kune network
   • Vanodzima zvisingakoshi
   • Vanoshandura mifananidzo kuita zvinyorwa
   • Vanoderedza saizi yedataset
6. Ndeipi modhi yakawedzera Pix2Pix kuti udzidze kubva kune isina kubatanidzwa mifananidzo?
   • ResNet
   • CycleGAN
   • 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