# IDSIA/sacred

Sacred is a tool to help you configure, organize, log and reproduce experiments developed at IDSIA.

Repository: https://github.com/IDSIA/sacred
Canonical: https://ross.abutalabs.com/products/sacred
Language: Python
License: MIT
License Family: permissive
Topics: python, machine-learning, infrastructure, reproducible-research, reproducibility, reproducible-science, mongodb
Last push: 2025-10-22T08:04:57+00:00

## Health v2 (maintenance only)
Score: 44/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 48, release rhythm 8, longevity 100
- inputs: {"age_days": 4538, "days_push": 315, "days_rel": 645, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4374, forks 393 (observed 2026-08-28T04:08:46.857816+00:00)

## What it is
Sacred is a Python library for configuring, organizing, logging, and reproducing machine learning and scientific experiments. It provides config scopes, dependency injection, a per-experiment CLI, observers that save run metadata to databases like MongoDB, and automatic random seeding.

## Use cases
- track parameters of machine learning experiments
- reproduce experiment results with fixed seeds
- log experiment runs to a MongoDB database
- run experiments with different settings from the command line
- record dependencies and machine info for each run
- organize and compare results of many experiment variants

## When to choose
- you run many Python experiments and need reproducibility and run tracking
- you want automatic config logging and CLI parameter overrides with minimal boilerplate
- you prefer a lightweight experiment tracker without a hosted service

## When to avoid
- you need a full-featured experiment dashboard with visualizations like W&B or MLflow
- your project is not in Python
- you need distributed training orchestration rather than experiment bookkeeping

## Facets
- artifact type: library
- maturity: maintenance
- function: configuration-management, logging, monitoring, cli, developer-tools
- domain: machine-learning, data-science, developer-tools
- platform: python, cli, cross-platform
- tags: experiment-tracking, reproducibility, machine-learning-infrastructure, observers, mongodb, automatic-seeding, automation

## Member repositories
- IDSIA/sacred (main) score 44

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:46.857816+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:21:20.503139+00:00, confidence not recorded.
  - readme: https://github.com/IDSIA/sacred (fetched 2026-08-28T04:08:46.857816+00:00, sha 8ff1cd63e0dd)
  - registry_pypi: https://pypi.org/pypi/sacred/json (fetched 2026-08-29T09:09:36.821955+00:00, sha 216e17e1e32a)
- Data as of 2026-08-30T08:39:29.467469+00:00.