Kedis ku xidhan hagaha · Dhexdhexaad Heerka

Kediska U-Net Architecture

Tijaabi U-Net encoder-decoder summetry, ka bood isku xidhka, qaybinta cilmiga noolaha, saadaasha qaylada faafinta, iyo beddelka

Wadooyinka haga ee la xidhiidhaDhismaha Net
Su'aal 1 ee 8

Maxaa siinaya U-Net qaabkeeda 'U'?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Maxaa siinaya U-Net qaabkeeda 'U'?
    • Cod-bixiye hoos u dhigaya tusaalaha iyo furaha summeerka oo kor u qaadaya
    • Iskudubbarid keli ah oo toosan oo isku dhafan
    • Laba shabakadood oo isbarbar socda oo aan lahayn wax xidhiidh ah
    • Wareegga soo noqnoqda ee tallaabooyinka waqtiga
  2. Waa maxay ujeedada isku xirka U-Net?
    • Si loo yareeyo tirada halbeegyada eber
    • Si aad tafaasiil fiican uga gudubto lakabyada codeeyaha si toos ah ilaa lakabyada dekoodhka ee iswaafaqaya
    • Si sawirka loogu beddelo miisaan cawl
    • Si aad gabi ahaanba uga gudubto tababarka
  3. Maxaa U-Net markii hore loogu talagalay?
    • Aqoonsiga hadalka
    • Saadaasha qiimaha saamiyada
    • Kala-qaybinta sawirka baayo-caafimaadka
    • Turjumaada qoraalka
  4. Gudaha Stable Diffusion, maxay U-Net ku saadaalisay tallaabo kasta?
    • Ciwaanka ugu dambeeya ee sawirka
    • Saamiga cadaadisku
    • Xagasha kamarada
    • Sawaxanka laga saarayo qarsoodiga buuqa badan
  5. Maxaa ku dhacaya macluumaadka goobta marka uu codeerku hoos u dhaco?
    • Waxay soo koobtaa 'waxa' jira laakiin waxay luminaysaa faahfaahinta 'meesha' saxda ah
    • Waxay helaysaa goobo pixel sax ah
    • Waxay u beddeshaa sawirka qoraal
    • Waxay labanlaabtaa xallinta
  6. Which augmentation did the original U-Net paper use to learn invariance from limited annotated biomedical images?
    • Replacing every image with a random label
    • Elastic deformations of the training images
    • Training only on unmodified test images
    • Removing the decoder path before training
  7. How did the original U-Net paper process large images beyond the network’s input size?
    • By discarding all image borders
    • By resizing every image to one pixel
    • By predicting on overlapping tiles and extrapolating context at tile borders
    • By running the model once on a text prompt
  8. What did the original U-Net paper’s weighted loss emphasize?
    • Only background pixels, regardless of distance
    • The most common image color
    • Noise prediction at a diffusion timestep
    • Border pixels between touching cells to help separate adjacent objects