Skip to main content

Small wrapper around submitit to simplify cluster submissions

Project description

resubmit

Small utility library to simplify job submission with Submitit on SLURM clusters.

Quick usage:

  • Install locally for development:
pip install -e .[debug]
  • Use in your project:
from resubmit import submit_jobs, maybe_attach_debugger

# attach remote debugger if requested
maybe_attach_debugger(args.get("port", None))

# submit jobs (list of dicts)
submit_jobs(jobs_list, my_entrypoint, timeout_min=60, block=True)

API

submit_jobs(...) 🔧

Submit multiple jobs to a Slurm cluster using Submitit.

Signature (short):

submit_jobs(jobs_args: Iterable[dict], func: Callable[[List[dict]], Any], *, timeout_min: int, cpus_per_task: int = 16, mem_gb: int = 64, num_gpus: int = 1, account: Optional[str] = None, folder: str = "logs/%j", block: bool = False, prompt: bool = True, local_run: bool = False, slurm_additional_parameters: Optional[Dict] = None, constraint: Optional[str] = None, reservation: Optional[str] = None)

  • jobs_args: iterable of per-job kwargs (each item is passed to func).
  • func: entrypoint called for each job (should accept a list or single job dict depending on your usage).
  • timeout_min, cpus_per_task, mem_gb, num_gpus: common Slurm resources.
  • account: optional Slurm account name.
  • folder: logs folder for Submitit files (supports %j for job id).
  • block: if True, waits for all jobs and returns results.
  • prompt: if True, asks for confirmation interactively; set to False for CI or tests.
  • local_run: run the jobs locally without Submitit (useful for debugging).
  • slurm_additional_parameters: pass any extra Slurm key/value pairs to Submitit.
  • constraint / reservation: cluster-specific options kept out of defaults — provide them explicitly if you need them (they take precedence over values in slurm_additional_parameters).

Example:

submit_jobs(
    jobs_list,
    my_entrypoint,
    timeout_min=60,
    num_gpus=2,
    prompt=False,
    constraint="gpu",
)

maybe_attach_debugger(port: Optional[int]) 🐞

Attach debugpy to the job when port is provided (> 0). Safe no-op if port is None or <= 0.

  • If debugpy (and submitit) are not available on the node, a RuntimeError is raised with an explanatory message.

Example:

# attach remote debugger only when a port is provided (e.g., from CLI args)
maybe_attach_debugger(args.get("port"))

Tips:

  • Use prompt=False when calling submit_jobs from scripts or CI to avoid interactive prompts.
  • Tests demonstrate non-interactive behavior (prompt=False) and optional constraint/reservation handling.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

resubmit-0.0.3.tar.gz (5.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

resubmit-0.0.3-py3-none-any.whl (5.3 kB view details)

Uploaded Python 3

File details

Details for the file resubmit-0.0.3.tar.gz.

File metadata

  • Download URL: resubmit-0.0.3.tar.gz
  • Upload date:
  • Size: 5.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for resubmit-0.0.3.tar.gz
Algorithm Hash digest
SHA256 61d545bf9bf0ed69c04718ac66cb261de43ac33e4d225c4c6cdaf214746de015
MD5 26c469c7d5b841e1f6a88cdbc6578718
BLAKE2b-256 dac6c67aefd8cbbc8505a0c69af2386f9a8a7819a06a2726059ba122860188b2

See more details on using hashes here.

File details

Details for the file resubmit-0.0.3-py3-none-any.whl.

File metadata

  • Download URL: resubmit-0.0.3-py3-none-any.whl
  • Upload date:
  • Size: 5.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for resubmit-0.0.3-py3-none-any.whl
Algorithm Hash digest
SHA256 12f6850a710fe101a6d264a8eb1d71be867393353828170f4c77292181e54e1e
MD5 92deec04818cd1e688fbb363265f2f2b
BLAKE2b-256 7d9d487f89f98952c1a6183aad03f616259cd94a62b2adbef5992ebcdd173d51

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page