stet

A Python library for making parameter sweeps safely resumable.

Stet is a Latin proofreading instruction meaning "let it stand" - written beside a correction that should be ignored. When stet sees a parameter combination it has already run, it does the same: leave it, it's done.

When a long-running experiment script is re-run, whether after a crash, a time limit, or deliberately to extend a sweep, stet automatically skips any parameter combinations that have already been completed.

import stet

@stet.once(store='markov_runs.csv', key=['alpha', 'n_states', 'seed'])
def solve_markov(alpha, n_states, seed, n_iter=10_000):
    # expensive computation
    ...

for alpha in alphas:
    for n_states in [10, 50, 100]:
        for seed in range(20):
            solve_markov(alpha=alpha, n_states=n_states, seed=seed)

On restart, any already-completed (alpha, n_states, seed) combinations are skipped:

[once] Skipping solve_markov(alpha=0.01, n_states=10, seed=0)
[once] Skipping solve_markov(alpha=0.01, n_states=10, seed=1)
...

Installation

uv add stet
# or: python -m pip install stet

With Parquet support:

uv add stet[parquet]
# or: python -m pip install stet[parquet]

Usage

Zero config: uses _stet_store.csv in the current directory, all parameters as the key:

@stet.once
def run_experiment(alpha, seed):
    ...

Named store: backend selected from file extension (.csv, .json, .sqlite, .parquet):

@stet.once(store='runs.sqlite')
def run_experiment(alpha, seed):
    ...

Key subset: only alpha and seed determine whether a run is skipped; n_iter is ignored:

@stet.once(store='runs.csv', key=['alpha', 'seed'])
def run_experiment(alpha, seed, n_iter=1000):
    ...

Utilities

stet.status()            # print a summary of completed runs
stet.reset()             # clear the store
stet.reset(key_dict={'alpha': '0.1', 'seed': '42'})  # remove one entry

Storage backends

Extension Backend Notes
.csv pandas CSV Default. Human-readable.
.json stdlib json No extra dependencies.
.sqlite stdlib sqlite3 Best for large stores and parallel workers.
.parquet pandas + pyarrow Requires stet[parquet].

Documentation

Full documentation including how-to guides, API reference, and explanation of design decisions: docs.

Key dates

PyPI data
Data sourced directly from PyPI's database.
  • Released:
Latest release

1 maintainer

PyPI data
Data sourced directly from PyPI's database.
Avatar for vinceknight from gravatar.com vinceknight

Credits

Author: Vince Knight

Requires

Python >=3.11

Provides Extra

parquet

Report project as malware

Download files

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

Source Distribution

stet-0.0.2.tar.gz (144.7 kB view details)

Uploaded Source

Built Distribution

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

stet-0.0.2-py3-none-any.whl (12.8 kB view details)

Uploaded Python 3

File details

Details for the file stet-0.0.2.tar.gz.

File metadata

  • Download URL: stet-0.0.2.tar.gz
  • Upload date:
  • Size: 144.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for stet-0.0.2.tar.gz
Algorithm Hash digest
SHA256 0263000b9ddeb61d18b00b8a22677d0321ada154349d731934a18016f0c2ae2b
MD5 24eeb7133a401e2573fd02a2da5520f3
BLAKE2b-256 9fc26c293359f203f9d0db69d8311c2af67fc00c8078185fa0a5f5fef8c1cc10

See more details on using hashes here.

Provenance

The following attestation bundles were made for stet-0.0.2.tar.gz:

Publisher: publish.yml on drvinceknight/stet

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file stet-0.0.2-py3-none-any.whl.

File metadata

  • Download URL: stet-0.0.2-py3-none-any.whl
  • Upload date:
  • Size: 12.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for stet-0.0.2-py3-none-any.whl
Algorithm Hash digest
SHA256 24d3bd62c01b6616afc661ca21c40b80733114d5fe4e0dbe7824d714e6a176f4
MD5 4a5c29d19525600ec42388933c926a3d
BLAKE2b-256 2ff7961d16c777b2c341fefbfda55224ef656aa26b1d0976a97c3aa88b8f4543

See more details on using hashes here.

Provenance

The following attestation bundles were made for stet-0.0.2-py3-none-any.whl:

Publisher: publish.yml on drvinceknight/stet

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

This release

0.0.2 This release

2 files

0.0.1

2 files

Supported by

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