Kedis ku xidhan hagaha · Dhexdhexaad Heerka

DDPM iyo DDIM Sampler Quiz

Is barbar dhig DDPM iyo DDIM muunada roga ah, oo ay ku jiraan tababbarka la wadaago, kontaroolada stochasticity, wakhti ka boodi, iyo qiyaasaha sawirka nadiifka ah.

Wadooyinka haga ee la xidhiidhaDdpm iyo ddim Sampler
Su'aal 1 ee 8

Waa maxay faa'iidada dhabta ah ee ugu weyn ee DDIM marka loo eego DDPM?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Waa maxay faa'iidada dhabta ah ee ugu weyn ee DDIM marka loo eego DDPM?
    • Waxay soo saari kartaa sawirro la mid ah tillaabooyin muunado oo aad uga yar
    • Waxay u baahan tahay shabakad tababaran oo gebi ahaanba ka duwan
    • Waxay ku darsataa qaylo badan oo random ah tallaabo kasta si loo helo tayada wanaagsan
    • Waxay kaliya ku shaqeysaa sawirada madow iyo caddaanka
  2. Muxuu habka neerfaha ee qaabka faafinta dhab ahaantii u bartaa inuu saadaaliyo inta lagu jiro tababarka?
    • Qoraalka ugu dambeeya ee sawirka
    • Sawaxanka lagu daray sawirka wakhti go'an
    • Saamiga isku-buufinta ee sawirka
    • Dejinta kamarada loo isticmaalay in sawirka lagu qaado
  3. Maxay DDIM u soo saari kartaa isla sawir isku mid ah oo ka soo baxay dhawaaq isku mid ah mar kasta?
    • Sababtoo ah waxay kaydisaa sawirro hore loo soo saaray
    • Sababtoo ah waxay had iyo jeer isticmaashaa degdeg ah
    • Sababtoo ah habka muunadeynta ayaa ah mid go'aamin kara iyada oo aan la isku durin randomity
    • Sababtoo ah waxay isku celcelis ahaan wadarta orod badan oo random ah
  4. Gudaha DDIM, waa maxay qiimaha halbeegga eta oo soo ceshanaya dhaqankii asalka ahaa ee DDPM?
    • eta = 0
    • eta = -1
    • eta = 100
    • ee = 1
  5. Qiyaastii immisa tillaabo ayuu DDPM-gii asalka ahaa caadi ahaan ugu baahnaa muunado tayo sare leh?
    • Boqolaal ilaa kun tallaabo
    • Dhab ahaan 3 tallaabo
    • Hal tallaabo
    • In ka badan hal milyan oo tallaabo
  6. Kuwan kee baa ah qiyaasaha tirada dhexe ee muhiimka ah ka hor intaanay hore u socon?
    • Cabbirka faylka wax soo saarka
    • Sawirka nadiifka ah ee la saadaaliyay x0
    • Dhererka gundhigga degdega ah ee isticmaalaha
    • Heerkulka GPU
  7. What does DDIM change compared with DDPM when using a model trained with the DDPM procedure?
    • It replaces the network with a text classifier
    • It changes the training data from images to audio
    • It changes the sampling trajectory while reusing the trained denoising network
    • It removes the need to estimate noise
  8. Why can DDIM skip selected diffusion timesteps during sampling?
    • The training process never added noise
    • Each skipped timestep is replaced by a larger image
    • It copies an earlier generated image from a cache
    • Its non-Markovian sampling path can connect noise levels without visiting every training step