Ikibazo gihuza ibibazo · Hagati Urwego

Ikibazo cya DDPM na DDIM

Gereranya DDPM na DDIM byerekana icyitegererezo, harimo amahugurwa asangiwe, kugenzura ibintu bidasanzwe, gusimbuka igihe, no kugereranya-shusho.

Inzira zijyanye no kuyoboraDdpm Na Ddim Samplers
Ikibazo 1 Bya 8

Ni izihe nyungu nyamukuru za DDIM kurenza DDPM?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Ni izihe nyungu nyamukuru za DDIM kurenza DDPM?
    • Irashobora gutanga amashusho agereranijwe murwego rwo hasi rwo gutoranya
    • Irasaba umuyoboro utandukanye rwose
    • Yongeraho urusaku rudasanzwe kuri buri ntambwe kugirango ube mwiza
    • Ikora gusa kumashusho-umukara-na-yera
  2. Ni ubuhe buryo bwo gukwirakwiza imiyoboro y’imitsi yiga guhanura mugihe cy'amahugurwa?
    • Ibisobanuro byanyuma kubishusho
    • Urusaku rwongewe kumashusho mugihe runaka
    • Ikigereranyo cyo kwikuramo ishusho
    • Igenamiterere rya kamera ryakoreshwaga mu gufata ifoto
  3. Kuki DDIM ishobora kubyara ishusho imwe kuva urusaku rutangira buri gihe?
    • Kuberako ikora amashusho yakozwe mbere
    • Kuberako burigihe ikoresha ikibazo gihamye
    • Kuberako uburyo bwo gutoranya ibintu ari deterministic nta guterwa inshinge
    • Kuberako igereranya byinshi bidasanzwe biruka hamwe
  4. Muri DDIM, niyihe gaciro ka parameter eta igarura imyitwarire yumwimerere ya DDPM?
    • eta = 0
    • eta = -1
    • eta = 100
    • eta = 1
  5. Hafi yintambwe zingahe DDPM yumwimerere yari ikeneye kuburugero rwiza?
    • Amajana gushika ku ntambwe igihumbi
    • Intambwe 3
    • Intambwe imwe
    • Intambwe zirenga miliyoni
  6. Niki muribi byingenzi byingenzi bigereranya DDIM igereranya mbere yo gutera imbere?
    • Ingano ya dosiye y'ibisohoka
    • Ishusho isukuye yahanuwe x0
    • Umukoresha yihuta gushyiramo uburebure
    • Ubushyuhe bwa 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