quiz bu lëkkale ak tegtal · Diggante Niveau

Quiz Convolution yuñ mëna tàqale ci xóotaayu

Jëfandikool xaaj bu xóotaayu ak poñ, méngale njëgu convolution, xayma misaal bu ndaw, te xàmmee jëmmal model mobile.

Yooni njiit yu jëm ciConvolution yuñ mëna tàqale ci xóotaayu
Laaj 1 de 8

Lan mooy ñaari jéego yi convolution buñ mëna tàqale ci xóotaayu xaaj convolution buñ miin?

Xoolaatal laaj bu nekk8 laaj

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

  1. Lan mooy ñaari jéego yi convolution buñ mëna tàqale ci xóotaayu xaaj convolution buñ miin?
    • Jéego buy boole, topp ci jéego bu normalisasioŋ
    • Ñaari convolusioŋ 3x3 yuñ dajale ak benn ReLU ci seen biir
    • Benn filtre spatial ci xóotaayu chaine bu nekk, ginaaw ga benn njaxasu chaine 1x1 ci pointwise
    • ab dekonvolusioŋ enkodeer ak ab dekonvolusioŋ dekodeer
  2. Ci jéego bu xóot bi, naka lañuy jappe chaine de input yi?
    • Chaine bu nekk dafay segg boppam ak kernel boppam
    • chaine yépp ñu ngi koy njëkka boole
    • Chaine yi deñu leen di jaxase ci anam wu mucc balaa ñuy segg
    • Chaine bu njëkk bi kese lañuy wax
  3. For many output channels, what operation-count reduction does the guide associate with a 3×3 depthwise separable convolution?
    • Lu tollu ci 2 yoon lu néew
    • Lu tollu ci 100 yoon lu néew
    • Dafay jëfandikoo xayma yu bari, du néew
    • Lu tollu ci 8 ba 9 yoon lu néew
  4. Lan mooy wareefu convolution pointwise (1x1) ci jëmmal bii?
    • Ngir wàññi dayo barab bi
    • ngir boole ay leeral ci chaine yi ci barab bu nekk
    • Ngir yokk enkodaasu position
    • Ngir dindi mbir yu ñaaw yooyu
  5. Ban architecture bu siiw moo siiwal convolution yuñ mëna tàqale ci xóotaayu aparey mobile yi?
    • Net mobile
    • Resnet
    • AlexNet
    • LSTM
  6. An implementation applies only the depthwise spatial filters and omits the pointwise layer. Which capability is missing from that block?
    • Spatial filtering within each channel
    • Separate kernels for the input channels
    • Learned combination of information across channels
    • Local spatial feature extraction
  7. Using the guide’s formulas with M = 4 input channels and N = 16 outputs, how many multiply-adds per location do standard and separable 3×3 convolutions require?
    • Standard 100; separable 576
    • Standard 576; separable 100
    • Standard 576; separable 36
    • Standard 64; separable 100
  8. In the guide’s cost ratio 1/N + 1/9 for a 3×3 separable convolution, what happens as the output-channel count N becomes very large?
    • The ratio approaches 9
    • The ratio approaches zero
    • The ratio becomes independent of kernel size for every kernel
    • The ratio approaches 1/9