Guide-yakabatana mibvunzo · Pakati Nhanho

Kufambira mberi Kukura kweGANs Quiz

Edzai kudzidziswa kweGAN, symmetric kukura, alpha fade-in, batch-variation features, mareti ekudzidza akaenzana, uye gare gare zvakagadziriswa zvivakwa.

Nzira dzekutungamira dzinoenderanaKufambira mberi Kukura KwemaGAN
Mubvunzo 1 ye 8

GAN inokura zvishoma nezvishoma inotanga sei kudzidziswa?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. GAN inokura zvishoma nezvishoma inotanga sei kudzidziswa?
    • Pakugadziriswa kwakazara kwechinangwa nekukurumidza
    • Pachisarudzo chakaderera chakadai se4x4 pixels
    • Nokudzidzira kusarura chete
    • Nechinyorwa chemashoko
  2. Ndeipi bhenefiti huru yekukura kugadzirisa zvishoma nezvishoma?
    • Inoita kuti kudzidziswa kugadzikane uye nekukurumidza kupfuura kudzidziswa pakugadziriswa kwepamusoro zvakananga
    • Inobvisa kudiwa kwejenareta
    • Inobvisa iyo dataset inodiwa
    • Inovimbisa zero artifacts
  3. Kana iyo nyowani yepamusoro-resolution layer yawedzerwa, inounzwa sei?
    • Pakarepo pasimba rakazara
    • Mukusarura chete, kwete jenareta
    • Zvakanyatsopera mukati nehuremu hwekubatanidza (alpha) inokwira kubva pa0 kuenda ku1
    • Nekudzidzira zvakare network kubva pakutanga
  4. Sei mitsara ichiwedzerwa kune zvese jenareta uye kusarura?
    • Nokuti jenareta chete haigoni kubudisa ruvara
    • Kuderedza saizi yedataset
    • Chete chezvinangwa zvekuona
    • Kuchengeta iwo maviri network symmetric uye yakaenzana sezvo resolution inowedzera
  5. Chii chinangwa cheiyo minibatch standard deviation layer yakaunzwa muProGAN?
    • Kukwidza mufananidzo wekupedzisira
    • Kumanikidza modhi
    • Kuti uwedzere kusiyanisa kwekubuda uye kuodza mwoyo mode kudonha
    • Kutara mifananidzo yekudzidziswa
  6. Chii chakaitika gare gare chakaderedza kudikanwa kwekukura kwakajeka kunofambira mberi?
    • Kugadzirwa kwemusarura
    • StyleGAN2 inoratidza yakagadziriswa dhizaini ine skip / yasara yekubatanidza inogona kuenderana nemhando
    • Kuchinjira kumifananidzo yegreyscale
    • Kubvisa iyo latent vector zvachose
  7. At alpha equal to zero during a ProGAN generator fade-in, which path supplies the blended output?
    • Only the newly added block at full strength
    • An unrelated pretrained text encoder
    • A random photograph taken from the test set
    • The previous lower-resolution path after upsampling
  8. What is the original ProGAN equalized-learning-rate method intended to balance?
    • The effective scale of learning across layers through dynamic weight scaling
    • The number of captions assigned to each face image
    • The timestamp of every generated training sample
    • The count of output file formats across batches