Idanwo ti o sopọ mọ itọsọna · Alabọde Ipele

U-Net Architecture adanwo

Test U-Net encoder-decoder symmetry, skip connections, biomedical segmentation, diffusion noise prediction, and alternatives.

Jẹmọ itọsọna awọn ọnaU Net Architecture
Ibeere 1 ti 8

Kini yoo fun U-Net ni apẹrẹ 'U' rẹ?

Awotẹlẹ gbogbo ibeere8 ibeere

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

  1. Kini yoo fun U-Net ni apẹrẹ 'U' rẹ?
    • Ayipada koodu ti o sọ silẹ ati oluyipada alamimu ti o ṣe agbega
    • A nikan ni gígùn akopọ ti convolutions
    • Awọn nẹtiwọki ti o jọra meji ti ko si asopọ
    • Loop loorekoore lori awọn igbesẹ akoko
  2. Kini idi ti awọn asopọ foo U-Net?
    • Lati din awọn nọmba ti sile si odo
    • Lati kọja awọn alaye aye to dara lati awọn fẹlẹfẹlẹ kooduopo taara si awọn fẹlẹfẹlẹ decoder ti o baamu
    • Lati yi aworan pada si iwọn grẹy
    • Lati fo ikẹkọ patapata
  3. Kini U-Net ti ṣe apẹrẹ fun akọkọ?
    • Idanimọ ọrọ
    • Asọtẹlẹ idiyele ọja
    • Pipin aworan biomedical
    • Itumọ ọrọ
  4. Ni Itankale Stable, kini U-Net ṣe asọtẹlẹ ni igbesẹ kọọkan?
    • Ifori ipari ti aworan naa
    • Iwọn funmorawon
    • Igun kamẹra
    • Ariwo lati yọ kuro ninu alariwo alariwo
  5. Kini yoo ṣẹlẹ si alaye aaye bi awọn apẹẹrẹ awọn koodu isale?
    • O ṣe arosọ 'kini' wa ṣugbọn o padanu alaye 'ibiti' kongẹ
    • O jèrè awọn ipo piksẹli deede
    • O yi aworan pada si ọrọ
    • O ṣe ilọpo meji ipinnu
  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