Harmonise CI runners with lecture-python.myst (RunsOn T4) once QuantEcon/meta#386 recommends a configuration

Aperta
  • Qualcosa non va?
#306 1 commento 0 reazioni 0 assegnatari Vedi su GitHub

I maintainer di solito rispondono entro 1 giorno

Nessuno ha ancora preso questa issue.

Valutazione

Difficoltà
4/5
Tempo stimato
3-5 giorni
Idoneità per principianti
35/100
Tipo di issue
Refactoring
Chiarezza
Abbastanza chiara
Stato di attività
Attiva
Stack tecnologico
aws, github-actions, python
Ambito
build-system, ci-cd

Direzione di ricerca

Start by reading QuantEcon/meta#386 for its runtime recommendation, then compare ci.yml, cache.yml, publish.yml and environment.yml with lecture-python.myst. If GPU is recommended, mirror its runner and JAX setup and validate with a cold cache.yml run, comparing lecture times and outputs against the 2026-10-22 CPU build; if CPU is recommended, record the evidence and close the issue.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

This edition builds on GitHub-hosted ubuntu-latest (CPU), while its English source, QuantEcon/lecture-python.myst, builds on RunsOn g4dn.2xlarge (Tesla T4) with jax[cuda13]. Harmonise the runners with the source, but only after the runtime investigation in QuantEcon/meta#386 recommends a configuration. On 2026-10-19 this edition's CPU cache build ran faster (73 minutes of notebook execution against 107), so moving to the GPU runner is not obviously a gain.

Workflow This repository lecture-python.myst
ci.yml, cache.yml, publish.yml runs-on: ubuntu-latest runs-on: "runs-on=${{ github.run_id }}/family=g4dn.2xlarge/image=quantecon_ubuntu2404/volume=80gb/spot=false"
JAX CPU jax from environment.yml pip install -U "jax[cuda13]", then scripts/test-jax-install.py and nvidia-smi

Until then the edition stays on CPU. Its outputs can differ from the English site in low-order digits for JAX lectures; that is expected, not a defect.

Tasks

  • QuantEcon/meta#386 concludes with a recommendation for this repository.
  • If it recommends the GPU runner: mirror the source's runner and JAX steps in ci.yml, cache.yml and publish.yml, including XLA_PYTHON_CLIENT_PREALLOCATE=false and whatever CPU pinning the source applies per lecture (see QuantEcon/meta#350); reconcile environment.yml's CPU jax. The push needs the workflow token scope.
  • Validate with a cold cache.yml run: every notebook executes, and the per-lecture times and outputs are compared with the CPU build of 2026-10-22 (39724402846).
  • If it recommends staying on CPU: record the decision and the evidence here, and shut the issue.

Related: #271 (the -W gate and warning-only cache failures, the cause of this edition's recent failed cache runs) is independent of the runner and is better fixed first.

Lingua principale
TeX
Stelle
5
Fork
4
Merge medio
1g 21h
PR unite (30g)
15

Preparare l'ambiente

Questo progetto non fornisce container di sviluppo, Dockerfile né guida per i contributori, quindi l'ambiente è a tuo carico: parti dal suo README e consulta la nostra guida al primo contributo per i passaggi generali.

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di QuantEcon/lecture-python.zh-cn

Tutte le issue di QuantEcon/lecture-python.zh-cn

Issue simili

Altre issue su Build System

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.