Maswali yanayounganishwa na mwongozo · Kati Kiwango

Maswali ya Majadiliano Yanayotenganishwa kwa Kina

Tekeleza mgawanyiko wa kina na kwa uhakika, linganisha gharama za ubadilishaji, ukokotoa mfano mdogo, na utambue muundo bora wa muundo wa simu.

Njia za mwongozo zinazohusianaConvolutions Kina Separable
Swali 1 ya 8

Je, ni hatua zipi mbili ambazo upitishaji unaoweza kutenganishwa kwa kina unagawanya ubadilishaji wa kawaida kuwa?

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  1. Je, ni hatua zipi mbili ambazo upitishaji unaoweza kutenganishwa kwa kina unagawanya ubadilishaji wa kawaida kuwa?
    • Hatua ya kuunganisha ikifuatiwa na hatua ya kuhalalisha
    • Mazungumzo mawili ya 3x3 yaliyorundikwa na ReLU kati yao
    • Kichujio cha anga cha kina kwa kila chaneli, kisha mchanganyiko wa mkondo wa 1x1
    • Ubadilishaji wa programu ya kusimba na uondoaji wa avkodare
  2. Katika hatua ya kina, njia za kuingiza data zinatibiwaje?
    • Kila chaneli huchujwa kwa kujitegemea na punje yake
    • Vituo vyote vinajumlishwa pamoja kwanza
    • Vituo vinachanganyika bila mpangilio kabla ya kuchujwa
    • Ni chaneli ya kwanza pekee ndiyo inayozungushwa
  3. For many output channels, what operation-count reduction does the guide associate with a 3×3 depthwise separable convolution?
    • Takriban mara 2 chini
    • Takriban mara 100 chini
    • Inatumia hesabu zaidi, sio chini
    • Karibu mara 8 hadi 9 chini
  4. Je, ni jukumu gani la ubadilishaji wa maana (1x1) katika muundo huu?
    • Ili kupunguza azimio la anga
    • Ili kuchanganya taarifa katika vituo katika kila eneo
    • Ili kuongeza usimbaji wa nafasi
    • Ili kuondoa masharti ya upendeleo
  5. Je, ni usanifu gani unaojulikana ulieneza mienendo inayoweza kutenganishwa kwa kina kwa vifaa vya rununu?
    • MobileNet
    • 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