Guide-yakabatana mibvunzo · Pakati Nhanho

Yakagadziriswa Pruning uye Layer Kudonhedza Quiz

Inotarisa kuti nei yakarongeka kuchekerera ichimhanyisa inference, kukosha kwemutsetse kunofungidzirwa sei, uye kuti kukwenenzvera kana kuti distillation kudzoreredza kunaka sei.

Nzira dzekutungamira dzinoenderanaYakagadziriswa Pruning Uye Layer Kudonha
Mubvunzo 1 ye 8

Sei kuchekerera kwakarongeka kuchiburitsa kukurumidza kukuru asi kuchekerera kusina kurongeka kazhinji kusingaite?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Sei kuchekerera kwakarongeka kuchiburitsa kukurumidza kukuru asi kuchekerera kusina kurongeka kazhinji kusingaite?
    • Structured pruning inobvisa mabhuroko akabatana anonyatso dzikisa matensor
    • Kuchekerera kusina kuumbwa kunowedzera saizi yemuenzaniso
    • Structured pruning inowedzera mamwe ma layer
    • Kuchekerera kusina kurongeka kunongoshanda paCPUs
  2. Chii chinonzi 'layer dropping' mumamiriro ekushandura?
    • Kuderedza mwero wekudzidza pachikamu
    • Kudzima mabhuroki ese ekushandura kudzikisa kudzika kwemuenzaniso
    • Kuwedzera dropout kune yese layer
    • Kuronga patsva zvidimbu zvisina tsarukano
  3. Ndeipi chiratidzo chinowanzo ratidza kuti transformer layer inogona kudonhedzwa zvakachengeteka?
    • Iyo layer ine maparameter akawanda
    • Iyo layer ndiyo yekutanga mune network
    • Iyo layer yekupinza uye yakabuda yakavanzika nyika dzakanyanya kufanana
    • Iyo layer inoshandisa yakanyanya ndangariro
  4. Ndeipi nhanho inowanzodzosa kurongeka mushure mekuchekerera kwakarongeka?
    • Kuwedzera batch size chete
    • Kudzokororazve kusvika ku16 bits
    • Kuwedzera kutarisisa misoro
    • Dunhu pfupi rekugadzirisa zvakanaka kana ruzivo distillation
  5. Misoro yetarisiro yega yega inowanzorongedzerwa sei pakuchekerera?
    • Nekuwedzera kwekurasikirwa kana musoro wakafukidzwa (sensitivity)
    • Nekurongeka kwearufabheti
    • Nechinzvimbo chavo mudura remazwi
    • Nehuwandu hwezviratidzo zvavakaona
  6. Ndechipi chikuru chinoshanda chekushandisa kuchekerera kwakarongeka kwakasanganiswa nedistillation?
    • Kugadzira data rekudzidziswa otomatiki
    • Kutora mhuri yediki, inokurumidza modhi kubva kune imwe hombe yekutarisa
    • Encrypting modhi uremu
    • Kuwedzera kureba kwehwindo remukati
  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