Kediska Dabagalka Tijaabada

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

1. Waa maxay ujeedada koowaad ee tijaabada tijaabada ee barashada mashiinka?
   • Si loo sameeyo orodyada kuwo dib loo soo saari karo oo la barbar dhigi karo iyadoo la duubayo xaaladdooda oo dhan
   • Si loo sameeyo moodooyinka si dhakhso leh ugu tababaran GPU-ga
   • Si otomaatig ah loo geeyo moodooyinka wax soo saarka
   • Si loo cadaadiyo miisaanka moodada kaydinta
2. Isku-darkee ayaa lama huraan u ah samaynta hal tabobar oo run ahaantii dib loo soo saari karo?
   • Kaliya lambarka saxda ah ee kama dambaysta ah
   • Nooca koodka, xog-ururinta, iyo hyperparameters
   • Heerkulka GPU-ga iyo xawaaraha taageeraha
   • Magaca saynisyahanka xogta
3. Midkee kuwan ka mid ah aaladda raadraaca tijaabada caanka ah?
   • Photoshop
   • Kubernetes
   • MLflow
   • PostgreSQL
4. Sidee bay inta badan raad-raacayaasha tijaabada ahi sida caadiga ah u qabtaan cabbirada inta lagu jiro tababarka?
   • Adigoo gacanta ku koobiyaynaya nambarada xaashida xaashida ka dib
   • Adigoo qaadaya shaashadaha terminalka
   • Adigoo akhrinaya koodhka isha ee moodeelka
   • Iyada oo la isticmaalayo API-ga wicitaanada la geliyo wareegga tababarka ee ku qulqulaya dhabarka
5. Aaway agabka waaweyn sida miisaanka moodeelka la badbaadiyay ee sida caadiga ah lagu kaydiyo nidaamka raadraaca?
   • Kaydinta shay/farshaxan, oo leh tixraacyo lagu hayo kaydka xogta badan
   • Gudaha safka xogta metrics
   • Xusuusta GPU-da si joogto ah
   • Dhammaantood lama kaydiyo
6. Sidee buu raadraaca tijaabada u kobcayaa xilliga moodooyinka luqadaha waaweyn?
   • Waa laga tagay sababtoo ah LLMs uma baahna dabagal
   • Waxa kale oo ay hadda qabtaa noocyada degdega ah, raadadka qiimaynta, iyo wax soo saarka tayada leh
   • Waxa kaliya oo ay la socotaa kharashaadka GPU hadda
   • Waxay gebi ahaanba ku biirtay software-ka tafatirka fiidiyowga
7. Why record a dataset version or content hash with an ML run?
   • To prove the model is unbiased
   • To guarantee identical GPU arithmetic
   • So the run can be linked to the exact data snapshot used
   • To replace the code version
8. In MLflow Tracking, where are large model files typically stored relative to parameters and metrics?
   • Only inside the metric names
   • Always embedded in the database row for each parameter
   • They are not recorded by a tracking system
   • In an artifact store, while run metadata is kept in a backend store