qs-dmss 0.9.0

Evidence-first simulation lab for the QuantumScalar Dark Matter Simulation Suite with Lab Mode, Campaign Studio study templates, decision profiles, dry-run Slurm request bundles, publication export composition, and reproducible evidence bundles.

QS-DMSS

QS-DMSS is a deterministic, evidence-first simulation lab for the QuantumScalar Dark Matter Simulation Suite.

The product loop is simple:

run simulations -> inspect evidence -> compare campaigns -> publish reproducible artifacts

QS-DMSS is not trying to be "just another solver." The project direction is to turn simulation runs into trustworthy research objects: configured, measured, bundled, verified, replayable, comparable, citable, and ready to share.

QS-DMSS is beta for reproducible package/evidence workflows; it is not peer-reviewed scientific validation.

Tangible Utility Summary

QS-DMSS is a classical, NumPy-first reference lab for small deterministic Schrodinger-Poisson-style quantum scalar dark matter experiments. Its strongest public lane is not high-performance cosmological discovery; it is making simulation studies fast to set up, easy to inspect, reproducible, comparable, and citable.

• Rapid sandbox studies for parameters such as the self-interaction term engine.g_int, timestep, packet width, amplitude, and random seed.
• Local Python package, CLI, cockpit, and JSON API paths that avoid HPC or cluster infrastructure for small reference runs.
• Evidence bundles, manifests, replay, verification, reports, and Zenodo/PyPI metadata that turn runs into portable research objects.
• Campaign Studio study templates for preserving, rerunning, importing, exporting, and explaining reproducible parameter-grid designs.
• Portable workspace snapshots for handing off selected runs, experiments, study templates, research-object exports, job provenance, collaborators, and annotations as local JSON.
• A packaged Self-Interaction Sweep study template focused on engine.g_int, with purpose, runtime target, metrics, limitations, non-claims, and guided interpretation visible in the cockpit before a user edits any YAML.

See docs/scientific-scope-and-utility.md for the scientific scope, non-claims, and a concrete self-interaction campaign study using engine.g_int.

• Installable Python package
• Bundled demo assets for installed-package smoke testing
• Config-driven simulation CLI
• Local-first run cockpit and JSON API
• Parameter sweeps and multi-run comparison in the cockpit
• Experiment registry with saved comparison reports and bundles
• Objective-driven decision profiles with ranked recommendations
• Template-defined decision campaigns across multi-parameter search grids
• Reusable Campaign Studio study templates for preserving, rerunning, and sharing campaign designs
• Run ledger with stable run IDs and config digests
• Evidence bundle with artifacts, metrics, manifest, and HTML report
• Replay and verification commands for reproducibility checks
• Canonical simulation showcase with CSV, SVG, report, run evidence, and replay evidence
• GitHub Actions CI and containerized runtime

What This Build Includes

The current reference implementation focuses on the backbone needed for an evidence-first simulation lab:

• A NumPy-based split-step Schrodinger-Poisson solver
• YAML configuration loading with explicit validation
• Structured run outputs under runs/<run_id>/
• Structured experiment outputs under experiments/<experiment_id>/
• A local cockpit for launch, inspection, verification, replay, and bundle download
• Sweep support for exploring one parameter across multiple deterministic runs
• Decision campaign support for expanding a template into a multi-parameter grid automatically
• Campaign Studio study templates that save, reopen, import, export, and display edited grids, scoring contracts, and last-run provenance
• A packaged Self-Interaction Sweep template that turns engine.g_int into a concrete tangible-utility demo after install
• Local workspace export/import for portable collaboration handoffs with collaborator and annotation metadata
• Dry-run Slurm request bundles that emit reviewable scheduler artifacts without submitting jobs
• Comparison tooling for energy drift, norm drift, density, and runtime deltas
• Decision profiles that score runs against an explicit objective, constraint set, and ranking policy
• Durable experiment exports with copied run evidence, comparison JSON, report HTML, manifest, and bundle ZIP
• LocalExecutor job records under jobs/<job_id>/job.json that preserve submitted config metadata, run/replay lifecycle state, multi-run campaign/comparison provenance, research-object export provenance, and returned artifact roles for future collaboration and HPC connector work
• Evidence artifacts:
  • config.yaml
  • run.json
  • metrics.json
  • energy.csv
  • environment.lock.json
  • artifacts/final_density.npy
  • artifacts/final_state.npz
  • report.html
  • manifest.sha256.json
  • evidence_bundle.zip
• Verification tooling for manifests and config digests
• Replay support for deterministic reruns

Quickstart

Install the published package from PyPI:

python -m venv .venv
.\.venv\Scripts\Activate.ps1
python -m pip install --upgrade pip
python -m pip install qs-dmss

Run the bundled demo config from the installed package:

qs-dmss run-demo

Launch the bundled installed-package demo campaign:

qs-dmss campaigns run-demo

Run the benchmark validation spine:

qs-dmss benchmarks validate --scenario demo-baseline

This writes benchmark-validation/benchmark-validation.json plus a human-readable benchmark-validation/benchmark-validation.md summary.

Run the canonical simulation showcase:

qs-dmss showcase run --output-root simulation-showcase

This writes a human-readable simulation-showcase/simulation-showcase.md walkthrough, CSV tables, SVG plots, verified run evidence, and replay evidence for the packaged canonical simulation scenario.

Generate a review-only Slurm request bundle without submitting to a scheduler:

qs-dmss executors slurm-dry-run configs/demo.yaml --request-root dry-run-jobs --job-name qs-demo

This writes job.json, request-bundle/request-bundle.json, request-bundle/slurm-job.sh, a copied config, and review instructions. The job state remains draft; QS-DMSS does not call sbatch.

For source development, install the checked-out repository in editable mode:

python -m pip install -e .[dev]
qs-dmss run configs/demo.yaml

Builders and sponsors can start with the product direction:

• docs/scientific-scope-and-utility.md
• docs/product-vision.md
• docs/contributor-roadmap.md
• docs/funding-roadmap.md
• docs/distributed-research-workspace-executor-architecture.md
• docs/slurm-site-policy-feedback.md

The current product milestone includes QS-DMSS Lab Mode plus an installable dry-run Slurm review target: a richer cockpit/showcase experience for running scenarios, inspecting outputs, comparing variants, verifying and replaying evidence, exporting polished research objects, and generating reviewable HPC request bundles that never submit jobs.

Public builder coordination now lives in issue #57. The latest Campaign Studio product slices on main add scenario metadata, editable parameter grids, decision-profile editing, scoring-contract preview, reusable study-template cards, and a packaged Self-Interaction Sweep template. That template gives fresh users one concrete engine.g_int campaign they can run, inspect, rerun, export, and critique without first designing a study from scratch.

Distributed collaboration and HPC connectors are possible future platform layers, but live collaboration and scheduler submission are not shipped runtime behavior yet. The current local-first seam supports portable workspace export/import with collaborators and annotations plus a dry-run Slurm request bundle generator that writes reviewable scheduler artifacts without calling sbatch, SSH, or a remote scheduler. This documents the path toward shared research workspaces, executor contracts, job lifecycle tracking, artifact collection, and scheduler guardrails. HPC administrators and research computing reviewers can use the Slurm site-policy feedback packet to review the generated bundle shape before any real submit/status/collect connector is attempted.

Review paths remain available for people who want to validate the public package:

• docs/reviewer-wheel-quickstart.md
• docs/reviewer-packet.md

Start the local cockpit:

qs-dmss cockpit --host 127.0.0.1 --port 8001

Then open http://127.0.0.1:8001 in a browser.

Inside the cockpit you can:

• Use Lab Mode to launch the packaged canonical simulation showcase, read guided interpretation, run a guided variant comparison, inspect the Evidence Explorer, preview generated reports/artifacts, compose a research object export, and open the full evidence outputs
• Inspect Scenario Library metadata for packaged scenarios, including purpose, expected runtime, artifacts, readiness, limitations, and suggested next actions
• Select the packaged Self-Interaction Sweep study template to inspect purpose, expected runtime, metrics, limitations, non-claims, and guided interpretation for an engine.g_int campaign
• Edit the Campaign Studio parameter grid and decision profile for the bundled decision campaign, preview the scoring contract, and launch the edited campaign through the existing evidence/recommendation workflow
• Save Campaign Studio edits as local study templates, inspect visible template cards with objective/run metadata, reload or rerun saved templates, and import/export the study JSON so another user can reproduce the same campaign design
• Inspect LocalExecutor job provenance for selected runs, campaign variants, saved experiment artifacts, and persisted research-object exports, including job ID, backend, lifecycle state, child jobs, and returned artifact roles
• Export or import a portable research workspace JSON with selected run, experiment, study-template, research-object, job, collaborator, and annotation metadata
• Generate a dry-run Slurm request bundle from a config for review before any manual HPC submission
• Launch a single run from a checked-in or edited config
• Launch a parameter sweep across interaction strength, timestep, step count, amplitude, width, or seed
• Launch a template-defined decision campaign that expands into a reproducible multi-parameter run matrix
• Compare multiple runs side by side with shared experiment metadata
• Save a comparison into the experiment registry and reopen it later with report and bundle downloads
• Load an objective-driven template and see the recommended winner directly in the comparison view

Verify the generated evidence bundle:

qs-dmss verify runs\<run_id>

Replay a prior run using the captured config:

qs-dmss replay runs\<run_id>

Persist a saved experiment bundle from two or more runs:

qs-dmss experiments export <run_id> <run_id> --label "comparison bundle"

List saved experiment artifacts:

qs-dmss experiments list

Launch the decision campaign defined by a template:

qs-dmss campaigns run configs/demo.yaml

Or launch the bundled installed-package demo campaign:

qs-dmss campaigns run-demo

The checked-in demo template now includes a decision profile:

• objective
• constraints
• ranking
• campaign

That means sweeps, experiment exports, and template-driven campaigns can now return a replayable recommendation instead of only raw metric tables.

The packaged showcase command adds a simulation inspection path on top of that loop:

run packaged scenario -> export CSV/SVG artifacts -> verify evidence -> replay -> compare final density

Container Runtime

Build the container image:

docker build -t qs-dmss .

Run the cockpit in Docker:

docker run --rm -p 8001:8001 qs-dmss

The image installs the built wheel, starts qs-dmss cockpit --host 0.0.0.0 --port 8001, and exposes the health endpoint at http://127.0.0.1:8001/api/health.

Project Layout

configs/                 Checked-in example configs
benchmarks/              Benchmark validation guidance
schemas/                 JSON schema for run configs
src/qs_dmss/             Package source
tests/                   Smoke and reproducibility tests
runs/                    Run ledger outputs (generated)
experiments/             Saved comparison artifacts (generated)

Development

Run the smoke tests:

pytest

CI lives in .github/workflows/ci.yml and validates:

• the editable install and test suite across Python 3.10 through 3.13
• static cockpit JavaScript syntax
• source distribution and wheel build metadata
• installed-wheel run-demo smoke test
• Docker build plus live /api/health and /api/configs probes

Fresh-install adoption smoke lives in .github/workflows/fresh-install-smoke.yml and validates PyPI and GitHub release-wheel installs on Linux, macOS, and Windows.

Release-candidate versioning and distribution artifact rules live in RELEASE.md.

PyPI distribution details and Trusted Publishing provenance live in docs/pypi-distribution-readiness.md.

The beta promotion gate lives in docs/beta-readiness.md.

Benchmark validation guidance lives in docs/benchmark-validation.md.

Canonical simulation showcase guidance lives in docs/simulation-showcase.md.

Evidence artifact definitions live in docs/evidence-bundle-glossary.md, demo and benchmark expectations live in docs/demo-benchmark-expectations.md, and decision profile fields are annotated in docs/decision-profile-example.md.

Product, funding, and builder-roadmap guidance lives in docs/product-vision.md, docs/funding-roadmap.md, and docs/contributor-roadmap.md.

Contributor source-map guidance lives in docs/contributor-map.md, and GitHub social preview setup lives in docs/social-preview.md.

Scholarly indexing readiness and public-launch materials live in docs/ascl-joss-readiness.md and docs/public-technical-launch-post.md.

The JOSS preflight checklist lives in docs/joss-preflight.md.

The active builder roadmap lives in docs/post-v0.3-active-roadmap.md.

The next research-grade upgrade slice is defined in docs/research-grade-upgrade-slice.md, with paper strategy notes in docs/research-paper-strategy.md.

Funding And Stewardship

QS-DMSS has been accepted into Open Source Collective. Support can be directed through Open Collective.

The current funding ask is concrete: help build QS-DMSS Lab Mode and the publication-grade artifact workflow around it. Funding should unlock visible public outcomes such as cockpit improvements, scenario packs, evidence exploration, report exports, campaign tooling, benchmark scenarios, and research-software documentation.

The funding roadmap lives in docs/funding-roadmap.md.

Funding support does not imply peer-reviewed scientific validation or endorsement of any physical model. Scientific claims should continue to be reviewed through reproducible evidence, public issues, and formal scholarly review.

Citation

Citation metadata lives in CITATION.cff. GitHub uses this file to populate the repository citation prompt, and Zenodo can use it when archiving GitHub releases.

For formal research references, prefer the Zenodo DOI citation:

• Project DOI: 10.5281/zenodo.20074924
• Latest archived release DOI: v0.8.0 / 10.5281/zenodo.20673804
• Previous archived release DOI: v0.7.0 / 10.5281/zenodo.20671389
• Previous archived release DOI: v0.6.1 / 10.5281/zenodo.20631860
• Previous archived release DOI: v0.6.0 / 10.5281/zenodo.20618884
• Earlier archived release DOI: v0.5.0 / 10.5281/zenodo.20617028
• Older archived release DOI: v0.4.0 / 10.5281/zenodo.20500433
• Prior archived release DOI: v0.3.0 / 10.5281/zenodo.20112923
• Older archived release DOI: v0.2.0 / 10.5281/zenodo.20091602
• Historical archived release DOI: v0.1.5 / 10.5281/zenodo.20076871
• First archived release DOI: v0.1.3 / 10.5281/zenodo.20074925
• Software Heritage archival: pending; add the SWHID after Software Heritage reports a completed archive for the release.

Zenodo citation notes live in docs/zenodo-citation.md.

Product Spine

QS-DMSS already has the package/evidence/reproducibility spine needed for a stronger product. Optional accelerator backends, plugin expansion, and broader research modules can build on a stable execution loop:

configure -> run -> measure -> bundle -> verify -> replay

The cockpit adds the first browser-native product layer on top of that loop:

configure -> launch -> inspect -> verify/replay -> compose research object

The experiment registry now makes comparison durable too:

select runs -> compare -> save -> report -> bundle -> reopen

The decision layer adds recommendation semantics to that flow:

select template -> launch campaign -> score runs -> recommend winner -> export evidence

The campaign layer now automates the search plan too:

select template -> expand campaign -> run matrix -> score variants -> recommend winner -> reopen bundle

Lab Mode turns that spine into a reviewer-facing simulation lab:

choose scenario -> run simulation -> inspect evidence -> compare variants -> verify/replay -> compose export