Ikibazo gihuza ibibazo · Hagati Urwego

Guhatira abarimu muburyo bukurikirana Ikibazo

Reba uburyo ubutaka-ukuri bwibanze butoza decoder ikurikirana, uburyo masike zitera gushyigikira imyitozo ibangikanye, nimpamvu umwanzuro utandukanye.

Inzira zijyanye no kuyoboraGuhatira abarimu muburyo bukurikiranye
Ikibazo 1 Bya 8

Mugihe cyo guhatira abarimu, niki kigaburirwa nkigitekerezo kuri buri ntambwe ya decoding?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Mugihe cyo guhatira abarimu, niki kigaburirwa nkigitekerezo kuri buri ntambwe ya decoding?
    • Ubutaka-ukuri ibimenyetso byabanje
    • Icyitegererezo ubwacyo cyahanuye
    • Ikimenyetso kidasanzwe
    • Ibisohoka byanyuma
  2. During teacher-forced training, what target is paired with decoder input y_(t−1)?
    • The previous ground-truth token
    • The next target token y_t
    • A token sampled from the validation set
    • The maximum-probability output from an earlier training run
  3. Guhuza ibizamini bya gari ya moshi biterwa no guhatira abarimu bizwi nka:
    • Kurimbuka
    • Uburyo bwo gusenyuka
    • Kubogama
    • Ikirango gisohoka
  4. Muri decoder ya Transformer, ni ubuhe buryo butuma amahugurwa ahatirwa na mwarimu akora mu buryo bumwe?
    • Kureka
    • Ibisanzwe
    • Gushakisha ibiti
    • Impamvu (reba-imbere) mask yo kwitondera
  5. Mugihe cyo gufata umwanzuro, kuki bidashoboka gukoreshwa mwarimu?
    • Hano nta bimenyetso-byukuri bizaza kugaburira
    • Igikorwa cyo gutakaza ntikiboneka
    • GPU iratinda cyane
    • Mask yo kwitondera irahagarikwa
  6. Ni ikihe gihombo gikoreshwa muri buri ntambwe mugihe cyo guhugura abarimu?
    • Gereranya ikosa ryakorewe kuri logi mbisi
    • Kwambukiranya-entropiya hagati yikwirakwizwa ryateganijwe nukuri gukurikira
    • Igihombo cya Hinge
    • Gutakaza Cosine
  7. The target sequence is “red, blue, green.” Under teacher forcing, which token is supplied when training the prediction of “green”?
    • The model’s prediction for “green”
    • The future token after “green”
    • The ground-truth previous token “blue”
    • A randomly sampled vocabulary item
  8. At decoder position t, which tokens must a causal attention mask block?
    • The source sequence
    • All earlier target tokens
    • The current position’s own input
    • Target tokens at positions later than t