Guide-yakabatana mibvunzo · Zviri nyore Nhanho

Data Augmentation Quiz

Review data augmentation methods for images and text, including label-preserving transformations, mixing and learned policies.

Nzira dzekutungamira dzinoenderanaData Kuwedzera
Mubvunzo 1 ye 8

Chii chinangwa chekutanga chekuwedzera data?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Chii chinangwa chekutanga chekuwedzera data?
    • Kuwedzera uye kusiyanisa data rekudzidzisa kuitira kuti mamodheru awedzere zvirinani
    • Kukurumidza kufungidzira pane yakadzidziswa modhi
    • Kudzvanya saizi yedataset pane dhisiki
    • Kuisa mazita emienzaniso isina kunyorwa otomatiki
  2. Ndeipi yeiyi nzira yakajairika yekuwedzera mufananidzo?
    • Gradient descent
    • Random horizontal flipping
    • Kukamura Mazwi
    • Dropout
  3. Ko shanduro yekumashure inoita sei pakuwedzera mavara?
    • Inodzosera kurongeka kwemazwi mumutsara
    • Inobvisa mazwi ese ekumisa kubva pachinyorwa
    • Inoshandura mavara kune mumwe mutauro uye kumashure kugadzira zvirevo
    • Inoshandura mavara kuita manhamba ekubatanidza
  4. Sei kuwedzera kuchifanira kunge kuri 'kuchengetedza label'?
    • Kuchengeta saizi yefaira redataset isingachinji
    • Kuita kuti kudzidziswa kumhanye nekukurumidza
    • Nekuti modhi haigone kuverenga mavara neimwe nzira
    • Nokuti kuchinja zita racho kwaizodzidzisa muenzaniso wacho kushamwaridzana kwakaipa
  5. Chii chinonzi Mixup technique?
    • Inosanganisa mienzaniso miviri uye mavara ayo kuita samuenzaniso yekudzidzisa
    • Inovhara nzvimbo ine rectangular yemufananidzo
    • Inotsvaga yakanakisa yekuwedzera mutemo otomatiki
    • Inowedzera ruzha rusina kurongeka kune uremu hwemuenzaniso
  6. Ndeipi nguva yekuwedzera data inonyanya kukosha?
    • Kana iwe uchinge uchinge uine unlimited yakanyorwa data
    • Kana data rakanyorwa riri shoma uye kuunganidza zvakawanda kwakaoma
    • Chete kana modhi yacho yanyatsodzidziswa
    • Chete chetabular spreadsheet data
  7. What does CutMix do to an image and its training label?
    • It adds a random word to a sentence without changing the label
    • It swaps only class names in the dataset
    • It replaces a patch with a patch from another image and mixes labels by the replaced area
    • It blurs every input and keeps a single label unchanged
  8. How does AutoAugment select augmentation policies in its original paper?
    • It chooses transformations from a fixed human-written list without evaluation
    • It uses test labels to relabel training images
    • It selects operations from inference-time confidence only
    • It searches for policies using validation performance as a reward signal