Maswali yanayounganishwa na mwongozo · Kati Kiwango

Ray kwa Maswali ya AI Iliyosambazwa

Angalia uelewa wako kuhusu kazi za Ray, watendaji (actors), marejeleo ya vitu, maktaba zilizosambazwa, na jinsi data ya vitu inavyosonga ndani ya nguzo (cluster).

Njia za mwongozo zinazohusianaRay Kwa AI Iliyosambazwa
Swali 1 ya 8

Ray anasuluhisha shida gani kimsingi?

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  1. Ray anasuluhisha shida gani kimsingi?
    • Kuongeza mzigo wa kazi wa Python na AI kutoka kwa mashine moja hadi kikundi kikubwa na mabadiliko madogo ya nambari
    • Kuweka lebo kwenye picha kiotomatiki
    • Inabana faili za mfano
    • Kubuni usanifu wa mtandao wa neva
  2. Katika Ray, kuna tofauti gani kati ya 'kazi' na 'mwigizaji'?
    • Majukumu ni ya picha na waigizaji ni wa maandishi
    • Kazi ni simu isiyo na utaifa ya kazi sambamba; mwigizaji ni mfanyakazi wa serikali ambaye anashikilia serikali
    • Wanafanana
    • Majukumu hutekelezwa kwenye CPU pekee na waigizaji kwenye vichapishi pekee
  3. Ray anarudi nini mara moja unapozindua kazi ya mbali?
    • Matokeo ya mwisho, daima
    • Ujumbe wa hitilafu
    • Wakati ujao (marejeleo ya kitu) ambayo utasuluhisha baadaye na ray.get()
    • Nakala ya nguzo nzima
  4. Ni maktaba gani ya Ray imeundwa kwa utaftaji wa hyperparameta iliyosambazwa?
    • Ray Kutumikia
    • Takwimu za Ray
    • RLlib
    • Ray Tune
  5. Which object type do Ray’s docs explicitly say can be read from same-node object-store memory without a copy?
    • A NumPy array stored in Ray’s object store
    • Any Python object whenever an ObjectRef is passed
    • An object requested from another node with no network transfer
    • Only actor handles stored in a separate database
  6. Je, ungetumia maktaba gani ya Ray kupeleka vielelezo nyuma ya sehemu ya mwisho ya kiotomatiki na inayoweza kubadilika?
    • Ray Tune
    • Ray Kutumikia
    • Treni ya Ray
    • Takwimu za Ray
  7. How does Ray Serve decide when to add replicas in its built-in request-driven autoscaling?
    • It reads the model’s training loss
    • It changes the number of model parameters
    • It monitors ongoing requests or queue demand against configured targets
    • It scales only when a human edits the deployment
  8. How does Ray move an object when a task on a different node needs it?
    • It shares the owner’s local memory directly across every node
    • It automatically copies all objects into each node’s GPU memory
    • It sends only an ObjectRef and never transfers the object data
    • It transfers the object over the cluster network; serialization, locality, and object size affect cost