Modèle quiz de potage

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

1. What does model pruning remove from a trained network?
   • Dindi ay poid wala ay jumtukaay yu amul benn njariñ ci génne gi
   • Yokk yeneen paramet ngir gëna dëggu
   • Taggat ab royukaay ngir yeneen jamono
   • Yokk dayo dugal
2. Lan mooy wuutale dagg buñ defar ak dagg buñ defarul?
   • Dagg buñ yamale du dox ludul ci nataal yi
   • Dagg buñ yamale dafay dindi yunit yu mat sëkk yu melni neuron wala bopp; amul yamale dindi poid yi benn-benn
   • Dagg bu jaxasoowul dafay laaj ñu tàggataat ko ci noonu
   • Amul benn wuute ci seen biir
3. Lan moo waral dagg garab gu jaxasoo du gaawaale model bi ci hardware bu gëna bari?
   • Dafay efaase ay diisaay yu bari
   • Saa yu nekk dafay wàññi njub ba amul dara
   • matrix yu ndaw yi deñu soxla ay aparey wala ay bibliothèque ngir mëna jëfandikoo ay zero
   • Dafay yokk limu parametre yi
4. Lan mooy rëset bu ñuy faral di def ngir dagg garab?
   • Dagg lépp benn yoon, ba noppi nga dugal ko
   • Liggéey bu njëkk, te bul musa tàggataat
   • efaase ay diisaay ci anam wu mucc ba njubte gi wàññeeku
   • Taggat, dagg parametre yi gëna néew solo, defar bu baax ngir defaraat, ba noppi baamtu
5. Luy xalaatu tiketu loteri bi wax?
   • Reseau yu dëgër yi deñuy am ay sous-reseau yu néew yu mëna méngoo ak model bi yépp suñu leen tàggatee ci initialisation bu dëggu bi
   • Dagg dafay gëna rafetal njub
   • Dagg lu bari mooy gëna baax
   • Mënuñu kompresse model yu mag yi
6. Ban tegtal bu yomb lañu faral di jëfandikoo ngir xam ban poid lañu wara dagg?
   • Tur yi ci alfabe
   • magnitude absolu bu poid yi
   • Kulòoru kàrtu tàmbali
   • Limu misaali tàggat yi
7. When is pruning most likely to reduce inference latency on ordinary dense hardware?
   • When individual zeros are added but the dense kernel still computes them
   • When the model is renamed without changing its graph
   • When whole channels or other structures are removed to produce a smaller dense model
   • When a sparse checkpoint is stored in a larger file
8. What should a team measure after pruning for a deployment target?
   • Only the number of zeros in a checkpoint
   • Only training accuracy before pruning
   • Whether the file extension changed
   • Task quality and latency or memory on the intended hardware