Ikibazo gihuza ibibazo · Hagati Urwego

Gutema no Gukata Ikibazo

Kugenzura impamvu gutema byihuta byihuta, uko akamaro kagereranijwe, nuburyo guhuza neza cyangwa gusibanganya bigarura ubuziranenge.

Inzira zijyanye no kuyoboraGutema no Gutema Ibice
Ikibazo 1 Bya 8

Ni ukubera iki gutema byubatswe bitanga umuvuduko nyawo mugihe gutema bitubatswe akenshi bidakora?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Ni ukubera iki gutema byubatswe bitanga umuvuduko nyawo mugihe gutema bitubatswe akenshi bidakora?
    • Gutema byubatswe bikuraho ibice bifatika bigabanya tensor
    • Gutema bitubatswe byongera ingano yicyitegererezo
    • Gutema byubatswe byongera izindi nzego
    • Gutema bitubatswe bikora gusa kuri CPU
  2. Niki 'guta ibice' murwego rwo guhindura ibintu?
    • Kugabanya igipimo cyo kwiga kuri buri cyiciro
    • Gusiba transformateur zose kugirango ugabanye ubujyakuzimu
    • Ongeraho guta ishuri kuri buri cyiciro
    • Gutondekanya ibice uko bishakiye
  3. Ni ikihe kimenyetso gikunze kwerekana ko urwego rwa transformateur rushobora kugabanuka neza?
    • Igice gifite ibipimo byinshi
    • Igice nicyambere murusobe
    • Igice cyinjiza nibisohoka byihishe birasa cyane
    • Igice gikoresha kwibuka cyane
  4. Niyihe ntambwe isanzwe igarura ubunyangamugayo nyuma yo gutema?
    • Kongera ingano yicyiciro gusa
    • Ongera ubare kuri 16 bit
    • Ongeraho imitwe myinshi
    • Uruziga rugufi rwo guhuza neza cyangwa gutandukanya ubumenyi
  5. Nigute imitwe yitabwaho kugiti cye ikunze gushyirwa mugukata?
    • Kwiyongera kw'igihombo iyo umutwe uhishe (sensitivite)
    • Ukurikije inyuguti
    • Ukurikije umwanya wabo mumagambo
    • Numubare wibimenyetso babonye
  6. Ni ubuhe buryo bukomeye bwo gukoresha ibiti byubatswe hamwe no gusibanganya?
    • Gukora amakuru yimyitozo mu buryo bwikora
    • Gukura umuryango wintangarugero ntoya, yihuta kuva kuri bariyeri nini
    • Shishoza uburemere bw'icyitegererezo
    • Kongera imiterere yuburebure bwidirishya
  7. Why can a smaller student recover some function of a pruned model during distillation?
    • It restores the deleted weights in place
    • It changes the test labels
    • It is trained to match useful outputs or behavior of the larger teacher
    • It removes the need for validation
  8. If a transformer layer’s input and output have high cosine similarity, what does that suggest under the guide’s pruning heuristic?
    • The layer is guaranteed to be useless for every task
    • The model must add another layer
    • The output has no features
    • The layer changed the representation relatively little and may be a candidate for dropping