Guide-yakabatana mibvunzo · Pakati Nhanho

Bottleneck Architectures Quiz

Inotarisa kudzikiswa kwechiteshi, kuverengera kuchengetedza, zvakasaririra zvakasaririra, yakaderera-chinzvimbo zvigadziriso uye yakagovaniswa dimensionality inomanikidza kuseri kwemabhodhoro.

Nzira dzekutungamira dzinoenderanaBottleneck Architectures
Mubvunzo 1 ye 8

MuResNet bottleneck block, ndeipi basa rekutanga 1x1 convolution?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. MuResNet bottleneck block, ndeipi basa rekutanga 1x1 convolution?
    • Kudzikisa huwandu hwematanho pamberi pe 3x3 convolution
    • Kuwedzera kugadzirisa kwenzvimbo
    • Kushandisa chikamu chekupedzisira
    • Kuwedzera ruzha rwekugadzirisa
  2. Sei bhodhoro richiita kuti matinji akadzika adhure kuverenga?
    • Inobvisa zvese activation
    • Inoita mabasa anodhura pamatanho mashoma
    • Inoshandisa mbesa hombe kwese kwese
    • Iyo inodarika backpropagation
  3. Mune autoencoder, iyo bottleneck layer inonyanya kusarudza:
    • Iyo optimizer sarudzo
    • Iyo mufananidzo faira fomati
    • Yakawanda sei iyo yekuisa inodzvanywa
    • Saizi yebatch
  4. MobileNetV2 inoshandisa ipi musiyano weiyo bottleneck zano?
    • Yakachena pooling bottleneck
    • Kudzokorora bhodhoro
    • Hapana bhodhoro zvachose
    • Inverted residual bottleneck (wedzera ipapo kondirakiti)
  5. LoRA inoshandisa sei misimboti yebhodhoro kumhando dzemitauro mikuru?
    • Nekuisa inodzidziswa yakaderera-nzvimbo (yakamanikana) bhodhoro muhuremu inogadziridza
    • Nekudzima matipi ekutarisisa
    • Nekupeta huremu hwese
    • Nekuverengera kusvika ku1 bit
  6. Iyo yakajairwa ResNet bottleneck block inotevera ndeipi chiteshi patani?
    • Yakapamhama, yakafara, yakafara
    • Deredza, gadzirisa, dzorera (semuenzaniso, 256 kusvika 64 kusvika 256)
    • Nhete, yakafara, nhete
    • Random per layer
  7. A ResNet block reduces 256 channels to 64 before its 3×3 convolution. Ignoring the 1×1 layers, what is the 3×3 layer’s approximate multiply-add cost relative to 256 channels?
    • About one-quarter
    • About one-half
    • About one-sixteenth
    • About four times as much
  8. An autoencoder uses a narrow latent code. What constraint does this impose on reconstruction?
    • The encoder must preserve every input pixel exactly
    • The network can no longer use learned weights
    • The decoder must expand the latent code before encoding
    • The decoder must reconstruct the input from a compressed representation