Guide-yakabatana mibvunzo · Pakati Nhanho

DDPM uye DDIM Samplers Quiz

Enzanisa DDPM neDDIM reverse sampling, kusanganisira kudzidziswa kwakagovaniswa, stochasticity control, kusvetuka-svetuka, uye fungidziro yemifananidzo yakachena.

Nzira dzekutungamira dzinoenderanaDdpm Uye Ddim Samplers
Mubvunzo 1 ye 8

Ndeipi mukana mukuru weDDIM pamusoro peDDPM?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Ndeipi mukana mukuru weDDIM pamusoro peDDPM?
    • Inogona kuburitsa mifananidzo yakafanana mumatanho mashoma ekuenzanisira
    • Inoda network yakadzidziswa yakasiyana zvachose
    • Iyo inowedzera yakawedzera ruzha ruzha pane imwe neimwe nhanho yemhando yepamusoro
    • Inongoshanda pamifananidzo yakasviba-uye-chena
  2. Chii chinonzi diffusion model's neural network chinonyatso dzidza kufanotaura panguva yekudzidziswa?
    • Chinyorwa chekupedzisira chemufananidzo
    • Ruzha rwakawedzerwa kumufananidzo panguva yakatarwa
    • The compression ratio yemufananidzo
    • Zvirongwa zvekamera zvakashandiswa kutora mufananidzo
  3. Sei DIM ichigona kuburitsa iwo chaiwo mufananidzo kubva kune imwecheteyo ruzha rwekutanga nguva dzese?
    • Nekuti inochengeta mifananidzo yakambogadzirwa
    • Nekuti inogara inoshandisa yakagadziriswa kukurumidza
    • Nekuti iyo sampling maitiro ndeye deterministic pasina jekiseni randomness
    • Nekuti inoverengera akawanda asina kujairika anomhanya pamwechete
  4. MuDDIM, ndeipi kukosha kweparameta eta inodzoreredza maitiro ekutanga stochastic DDPM?
    • uye = 0
    • eta = -1
    • uye = 100
    • uye = 1
  5. Angangoita matanho mangani ayo yekutanga DDPM yaiwanzoda masampuli emhando yepamusoro?
    • Mazana kusvika kuchuru nhanho
    • Manhanho matatu chaiwo
    • Nhanho imwe chete
    • Matanho anopfuura miriyoni imwe
  6. Ndeipi yeiyi ndiyo yakakosha yepakati huwandu hwekufungidzira kweDDIM usati waenda kumberi?
    • Saizi yefaira yezvakabuda
    • Iyo yakafanotaurwa yakachena mufananidzo x0
    • Kureba kwekumisikidza kwemushandisi
    • Iyo tembiricha yeGPU
  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