Kedis ku xidhan hagaha · Dhexdhexaad Heerka

Kediska Kediska Kediska ee GigaGAN

Sahami qaab dhismeedka qoraalka leh ee GigaGAN, xawaaraha jiilka, takoorida cabirka badan, tafatirka daahsoon, iyo kor u qaadida xallinta sare ee goonida ah.

Wadooyinka haga ee la xidhiidhaKoronto-dhaliyeyaasha Miisaaman ee Gigagan
Su'aal 1 ee 8

What total parameter count did the paper report for one high-capacity GigaGAN framework configuration?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. What total parameter count did the paper report for one high-capacity GigaGAN framework configuration?
    • 1.0 billion total across the 652.5M base generator and 359.1M upsampler
    • 70 million total across a 64M generator and 6M discriminator
    • 1.0 billion in the base generator alone, with no upsampler
    • 3.5 billion total across two text encoders
  2. Waa maxay faa'iidada xawaaraha weyn ee GigaGAN marka loo eego moodooyinka faafinta?
    • Uma baahna GPU
    • Waxay soo saartaa sawir hal baas oo hore ah halkii ay ka ahayd tallaabooyin badan oo beenin kara
    • Waxay ku tababartaa hal ilbiriqsi gudaheed
    • Ma isticmaasho shabakad neural ah
  3. Maxay qabtaa GigaGAN's 'muunad-doorashada kernel-ka'?
    • Wuxuu tirtiraa pixels aan la isticmaalin
    • Ku cadaadi sawirka JPEG
    • Wuxuu isku daraa filtarrada isdhaafsiga bangi isagoo isticmaalaya miisaanyo laga soo qaatay isla qoraalka
    • Si otomaatig ah ayuu u dooranayaa GPU-ga
  4. Sidee buu GigaGAN ugu daraa macluumaadka qoraalka jiilka sawirka?
    • Adigoo goosanaya degdega
    • Gabi ahaanba way iska indhatirtay qoraalka
    • Adigoo daabacaya qoraalka sawirka
    • Iyada oo loo marayo feejignaan isdhaafsan oo ku saabsan gundhigga qoraalka iyo is waafajinta sifooyinka CLIP
  5. Awooddee dheeraad ah ayuu GigaGAN bixiyaa ka baxsan jiilka aasaasiga ah?
    • Sameeyaha gaarka ah ee GAN kaas oo gaari kara xallinta 4K
    • Abuuritaanka maqalka
    • Aqoonsiga hadalka
    • Weydiinta database
  6. Waa maxay sababta meesha qarsoon ee GigaGAN faa'iido u leedahay tafatirka agabka?
    • Lama tafatirin karo
    • Waxay taageertaa isdhexgalka siman iyo isku dhafka qaabka isbeddellada la xakameyn karo
    • Waxa ay kaydisaa qoraalka oo kaliya
    • Waxay tirtirtaa sawirka hal isticmaal ka dib
  7. How does GigaGAN’s image discriminator make predictions across resolutions?
    • It scores only the text prompt and ignores generated images
    • It predicts one label at the final resolution only
    • It receives an image pyramid and predicts real or fake at multiple input and downsampling scales
    • It compares generated images with a fixed set of training captions only
  8. What does GigaGAN’s base generator produce for later multi-scale processing?
    • A single class label for the entire image
    • A sequence of text tokens for an autoregressive decoder
    • Only the final upsampled image, with no intermediate scales
    • An image pyramid with outputs at several resolutions