51-Pipeline Card Template v1.0 | Chapter 12 — Outputs & Release (Directory / Visualization / Comparative Scoring)

Home ／ Docs-Technical WhitePaper ／ 51-Pipeline Card Template v1.0

Chapter 12 — Outputs & Release (Directory / Visualization / Comparative Scoring)

I. Purpose & Scope

• Standardize the directory layout, visuals & tables, and comparative scoring (vs. baseline/version-diff) for pipeline deliverables, ensuring auditability, reproducibility, and comparability across teams.
• For outputs involving path quantities (arrival time/phase), the text must explicitly show gamma(ell) and d ell, and the data side records delta_form ∈ {general, factored}; publication requires p_dim = 1.0.

II. Prerequisites & Inputs

• Validation pre-checks: /validate passes G1–G8 with no S1–S5; check_dim_report.json passes (p_dim = 1.0).
• Contract consistency: stage.yaml, contract.yaml/schema.json, and Parameter/Error cards (cov_group/coverage) consistent with this volume.
• Timebase & sync: Chapter 5 thresholds satisfied (clock_state="locked", σ_y(τ), |δt_abs|/|Δτ_ch| within limits).
• Path consistency: len(gamma_ell)=len(d_ell)=len(n_eff)≥2, Δell compliant; phase computations aligned to the reference window.

III. Artifacts & Directory

1. Required:
   • Figures: DAG/state, timebase & sync health, path profiles & intervals, performance (latency/throughput/resources), comparative scoring.
   • Tables: compliance (G1–G8), key metrics & intervals, scoring details, inbound/outbound contract snapshots.
   • Reports: check_dim_report.json, validate_report.json, perf_summary.md, audit.jsonl, report_manifest.yaml, signature SIGNATURE.asc.
2. Units & dimensions: all numeric outputs must show units; any expression with division/integrals/composites must use parentheses.

Recommended layout

PTN_EXPORT/

figs/

pipeline_dag.svg

state_timeline.pdf

sync_health.pdf

path_profile.pdf

latency_hist.pdf

throughput_series.svg

scorecard_bar.pdf

tables/

kpi_summary.csv

compliance_table.csv

scorecard.csv

contracts_snapshot.csv

reports/

check_dim_report.json

validate_report.json

perf_summary.md

audit.jsonl

manifests/

report_manifest.yaml

SIGNATURE.asc

IV. Visualization Standards

1. Formats: dual export for each figure (vector PDF/SVG and bitmap PNG/JPG), DPI ≥ 300; axes with explicit units (s, rad, 1, m/s, MB/s, etc.).
2. Captions: include see[]/version; for path figures, annotate Δell and delta_form.
3. Minimal figure set:
   • DAG/state: pipeline_dag.svg, state_timeline.pdf.
   • Sync health: sync_health.pdf (time series with thresholds for δt_abs/Δτ_ch/σ_y(τ)).
   • Path profile: path_profile.pdf (n_eff(ell) with T_arr/Phi interval bands).
   • Performance: latency_hist.pdf, throughput_series.svg, resource_util.pdf.
   • Scoring: scorecard_bar.pdf (weighted total and per-metric vs baseline).

V. Comparative Scoring

1. Metrics & mapping:
   • Primary: ΔT_arr (s), r_phi (1), ε_flux (1), p_dim (1), Q_res (1).
   • Normalization: z_m = ( m − m_baseline ) / σ_baseline.
   • Scoring: q_m = 1 / ( 1 + exp( a z_m + b ) ) (default a=1,b=0; flip sign if “larger is better”).
   • Aggregate: Q = ( ∑_i w_i q_{m_i} ) / ( ∑_i w_i ).
2. Decision thresholds (aligned with Ch. 9/10):
   • Positive: core gates pass (|ΔT_arr| + U(T_arr) ≤ τ_T, LB(r_phi) ≥ r_phi_min, p_dim = 1.0, P95(ε_flux) ≤ guard) and Q ≥ Q_base + δQ_min.
   • Otherwise: Negative / [Restricted].
3. Publication tables: record computation conventions, weights w_i, intervals, and threshold comparisons in both scorecard.csv and scorecard.json.

VI. Compliance Block

• Display: p_dim, anchor_coverage/ver_presence, clock_state/τ_calib, coverage conventions (k/alpha/quantile), covariance PD check, G1–G8 pass rates and stops_triggered.
• If any core gate fails, tag [Restricted] across figures and tables and publish qualitative trends & diagnostics only.

VII. Machine-Readable Configs
A. report_manifest.yaml

version: "1.0.0"

bundle:

figs:

- "figs/pipeline_dag.svg"

- "figs/state_timeline.pdf"

- "figs/sync_health.pdf"

- "figs/path_profile.pdf"

- "figs/latency_hist.pdf"

- "figs/throughput_series.svg"

- "figs/scorecard_bar.pdf"

tables:

- "tables/kpi_summary.csv"

- "tables/compliance_table.csv"

- "tables/scorecard.csv"

- "tables/contracts_snapshot.csv"

reports:

- "reports/check_dim_report.json"

- "reports/validate_report.json"

- "reports/perf_summary.md"

- "reports/audit.jsonl"

metadata:

pipeline_id: "ptn-core"

method_version: "2.0.0"

created_at: "2025-09-24T16:00:00Z"

checksums: ["sha256:..."]

see:

- "EFT.WP.Core.Equations v1.1:S20-1"

- "EFT.WP.Core.Metrology v1.0:check_dim"

B. scorecard.json (example)

{

"version": "1.0.0",

"baseline": { "id": "base-001", "Q": 0.62 },

"method": { "id": "pipeline-A", "Q": 0.78 },

"weights": { "DeltaT_arr_s": 0.35, "r_phi": 0.25, "epsilon_flux": 0.15, "p_dim": 0.15, "Q_res": 0.10 },

"metrics": {

"DeltaT_arr_s": { "mean": -2.3e-9, "std": 4.8e-9, "U_k2": 1.5e-9 },

"r_phi": { "value": 0.72, "lb95": 0.61, "ub95": 0.80 },

"epsilon_flux": { "median": 0.004, "p95": 0.011 },

"p_dim": 1.0,

"Q_res": 0.13

},

"decision": "pass",

"see": ["EFT.WP.Core.Equations v1.1:S20-1","Data.Benchmarks v1.0:PROTO"]

}

C. kpi_summary.csv (headers)

stage,latency_p50_s,latency_p95_s,throughput_rps,rho_util,q_res,p_dim,epsilon_flux,allan_1s

ingest,0.032,0.085,1200,0.68,0.14,1.0,.,1.1e-11

arrival,0.047,0.110,950,0.72,0.13,1.0,0.006,1.2e-11

VIII. Anti-Patterns & Fixes

• Anti: reporting means without intervals → Fix: add U = k·u_c or quantile bands and state coverage mode.
• Anti: path figure missing delta_form/Δell → Fix: add in caption and align with n_eff.
• Anti: T_arr = ∫ n_eff / c_ref d ell (missing parentheses) → Fix: T_arr = ( ∫ ( n_eff / c_ref ) d ell ).
• Anti: publishing without check_dim_report.json → Fix: validate first, ensure p_dim = 1.0.

IX. Validation & Release Flow

• Generate & verify: run /validate, check_dim_report.json, performance & scoring tables/figures.
• Archive & sign: write artifacts to PTN_EXPORT/ and produce report_manifest.yaml and SIGNATURE.asc.
• Audit & lock: audit.jsonl records versions & checksums; lock versions and set rollback points as needed.
• External release: cite v1.* only; anchor coverage ≥ 90%; tag [Restricted] where non-compliant.

X. Cross-References

• Architecture & computation graph (Ch. 3); stage specs (Ch. 6); resources & performance (Ch. 8); quality gates & monitoring (Ch. 9); UQ coupling & feedback (Ch. 10).
• Parameter Card: data contracts & interfaces (respective chapters); Error Budget Card: scoring & thresholds (Ch. 8/Ch. 9).

XI. Checklist

• Dual figure exports (PDF/SVG and PNG); axis units and see[]/version in captions; path figures note Δell and delta_form.
• scorecard.csv/json consistent with baseline/threshold comparisons; weights, intervals, and decisions transparent.
• compliance_table.csv, kpi_summary.csv, contracts_snapshot.csv generated with complete units/dimensions.
• check_dim_report.json, validate_report.json, perf_summary.md, audit.jsonl, report_manifest.yaml, and signature complete.
• /validate passed with no S1–S5; if Restricted, all outputs tagged [Restricted] and only qualitative statements included.
• Citations compliant (volume+version+anchor, coverage ≥ 90%), no external links/aliases; versions & checksums consistent.