Seattle Building Permit Completion
Will a newly issued Seattle building permit for a substantive construction project reach final inspection sign-off within four years of issuance?
Versioned research data. Read the suitability requirements before interpreting scores.
Suitability and evaluation
Evaluation setting: retrospective research. Read the prediction setting and limitations below. Public validation is development data; downloadable holdout labels must stay outside the evaluated agent workspace.
At a glance
| Question | Will a newly issued Seattle building permit for a substantive construction project reach final inspection sign-off within four years of issuance? |
|---|---|
| Field | City of Seattle - Seattle Department of Construction and Inspections (SDCI), published on data.seattle.gov / binary-classification |
| Rows | 24,747 |
| Documented splits | 3 |
| This version | v-67a46c698fa813cc |
| Released | 2026-11-15 |
| Validation ROC AUC | 0.635545439690134 |
| Validation Brier | 0.17718219376270802 |
| Artifact | a220e68c960ee6b6aa473f015112be0993353ab3d12f9820c31551164f0ee677 |
Get this dataset
How to load it
from datasets import load_dataset
ds = load_dataset('earino/factory-seattle-building-permit-completion', revision='v-67a46c698fa813cc')
The version links and checksums identify this artifact. Exclude the labelled holdout from the agent's workspace when running an evaluation.
What is inside 
| Split | Rows | Positive labels |
|---|---|---|
| eval | 7,135 | — |
| holdout | 6,326 | — |
| train | 11,286 | — |
Files and checksums
| Download | Bytes | SHA-256 |
|---|---|---|
| LICENSE-EVIDENCE.json | 717 | ff7141e98807cf0373bd8c0a0bce28cbc0bb0409f80211778639668f1ed148b0 |
| LICENSE.md | 1,202 | f33b585718c4e0cf2c51e5e34184b41df83b074f0b932495b11f2b107a660477 |
| README.md | 6,886 | 07dae93ceefcb4e1155cf52e67d3f4fa1becb0d45d67bf4e4ba8b939d75585ef |
| SOURCES.json | 1,477 | 04758e5354cc9d2eaf154a77f3d2ba0225b00bf3fd2d65c0f382a57220998c0d |
| evaluation.json | 1,846 | c81d73707f4851236dacfd8752c741be5a79d089d40cbfe02a715a5abf90e379 |
| extract__meta.json | 2,081 | 767e29ef2df31a06b97bf83f62027fe7a40e376bf3b98ce238b79e7a5c6acf68 |
| extract__private__holdout.csv | 1,056,984 | aea92d0461c05b38c321e7012138c494921c01525676fb781e0fbb2885e44bd4 |
| extract__public__eval.csv | 1,180,318 | db5726b6c72f69815bade3f012146a198b8772589f255bccb3af094c5d076017 |
| extract__public__train.csv | 1,848,274 | 525fdf9a69b3449654fecebcc0087e1155c1fe00a4de72175cbf7366566da283 |
| extract__quality.json | 30,671 | 80980921fbb4851537bf2c2c6dabd2bc237bb17033eb224e643117f1bd7fafe6 |
| get_dataset.py | 3,680 | fad89222d50f9d9709225c34909be74c591df409c08b8b5951eba9d3cadfdd48 |
| qualification.json | 18,184 | 35bee5d4c3288600aa2e88c98fb2575a460299a74ae4e56c2f8a0630c8639f78 |
| reproduction__LICENSE-CC0-1.0.txt | 7,048 | a2010f343487d3f7618affe54f789f5487602331c0a8d03f49e9a7c547cf0499 |
| reproduction__LICENSE-MIT.txt | 1,124 | 56b932ff064e5da113ae10175c8546325f64f26e1ece71683ffcbf8b2304fe1c |
| reproduction__artifact_scrub.py | 11,369 | 8c663253e691a7b8751f9c853b40a3752a833360caafb1ea8e8f534e64af4b68 |
| reproduction__build.py | 17,093 | 38de47212ea79f8afe764f4b359bcc6a41f3343620dc6e5c7d484e78d76a3d37 |
| reproduction__construction.json | 30,615 | 75cf2be1eff93324e133dca424efecf1ff0f67a0037500362d59cab4601d4420 |
| reproduction__dataset_package.py | 7,927 | c2364defe58dcc99322be59a1bb04894ce00e823ce37afeb37bc31012f744327 |
| reproduction__dataset_run.py | 12,381 | 8085fae14d6e93b852e692e18c1ef72113a3601f48f78901b6d1ae6db278de57 |
| reproduction__get_dataset.py | 3,680 | fad89222d50f9d9709225c34909be74c591df409c08b8b5951eba9d3cadfdd48 |
| reproduction__leakage_battery.py | 17,379 | adbf7ee0aca0012b78cc1f1e4aeb968a23717cae30f66a02e080215e75f830e3 |
| reproduction__provenance.json | 573 | 11a9c4f355a6345b1a64b1bb34afd669daa68a62b53b6635c45f9861295a81a7 |
| reproduction__qualify_dataset.py | 66,067 | bff6d43d0904182e2612c32bab702e41addbf389904a236c209e78be39e37a6b |
| reproduction__recovery_recipes.py | 17,609 | f64c542fcda232169c95ace17421b45275194b934938c0ea4c9b68e6f844ad0e |
| reproduction__requirements.txt | 160 | 5c72f3268174d20364624f76680e189c408a05dba2e7b8cfe12d077b3865e7f0 |
| reproduction__source.json | 26,699 | fd603570c6993ec08a43a4f940631c7976c63a268e798aab339f831c5dd3557b |
| reproduction__source_io.py | 5,607 | 0c60eed4968cd342148efbaac7e4fa98f00708d8e6c14c0929928687b9af1ea3 |
| reproduction__train.py | 2,500 | a3c6bcf13735bc85c52129ded68f839090dffdc266ebc3810dc61b2e6ea5e7e8 |
| reproduction__validate.py | 1,277 | b597e7f84fed614e64b4a86fbecbd6ec0145916a24d0d221a796586976418e7b |
| reproduction__validate.sh | 482 | 3f06ca48f11c2e05d331b4d8394284925de660c1d4b086d2f8243959e83b7e6d |
| source-checks.json | 11,833 | 558d27301de4388608218b8445900620c0b13dedce36603a1f998252ceea4f29 |
| source-snapshot.tar.gz | 1,471,652 | 78cb100e796ccd05292a3fabe93b1c1ba7d13753bb694553f5a71fd3f5cfa0a7 |
| verify_dataset.py | 1,276 | f4a4bb7f565cdbd075c520621f881876b491eaed91a5e6e70afb31e4753f9834 |
Release notes · 2026-11-15
Will a newly issued Seattle building permit for a substantive construction project reach final inspection sign-off within four years of issuance?
Population: Every row in the live Socrata snapshot with PermitTypeMapped equal to Building, IssuedDate in the half-open window 2016-01-01 through 2022-10-01, and EstProjectCost at least 50000 nominal USD. Rows with impossible chronology (CompletedDate before IssuedDate, or IssuedDate before AppliedDate when both are present) are excluded as a source data-quality rule per the construction recipe. The unit is the permit as issued; no site-level or contractor-level aggregates are used. The population is fixed by permit type, issuance date and declared cost, not by the label value.
Target: y = 1 if CompletedDate is non-null and CompletedDate <= IssuedDate + 48 calendar months (dateutil relativedelta, month arithmetic); y = 0 otherwise, including permits still under inspection, expired permits and withdrawn permits. Evaluated against the 2026-11-14 snapshot, by which date the 48-month window of every eligible cohort has already closed, so the label can no longer flip from 0 to 1.
Prediction time: The permit IssuedDate, when the permit is issued. Only fields written at submittal, during plan review, or at issuance are used; CompletedDate, StatusCurrent, ExpiresDate and post-issuance revisions are excluded from features. The label is observed retrospectively after the 48-month window.
Splits: Fixed calendar boundaries on IssuedDate, no shuffling: train = 2016-01-01..2018-12-31 (11,289 eligible permits), eval = 2019-01-01..2020-12-31 (7,135), holdout = 2021-01-01..2022-09-30 (6,326). The holdout end date is chosen so the last 48-month label window closes on 2026-09-30, before the 2026-11-14 snapshot. No embargo is applied or needed: every feature is strictly row-local, there are no lags, rolling aggregates or other permits' outcomes, so moving a boundary by a month cannot leak anything.
Outcome horizon: 48 months from IssuedDate
Inputs: permitclass: Project class recorded on the application: Single Family/Duplex, Commercial, Multifamily, Institutional, Industrial, Vacant Land, or N/A. Categorical and row-local.; permitclassmapped: Residential, Non-Residential or N/A roll-up of permit class.; permittypedesc: New, Addition/Alteration, Tenant Improvement, Temporary, Change of Use Only - No Construction, Curb Cut, or NONE.; estprojectcost: Applicant-declared fair-market construction cost in nominal USD; all included rows have at least 50000 by population definition. Mutable if the project is later modified, so read as current-vintage.; housingunitsadded: Housing units the project adds; null for many rows and left blank, not zero.; housingunitsremoved: Housing units the project removes; null left blank.; housingunits: Net housing-unit change for the project; null left blank.; applied_to_issued_days: Calendar days from AppliedDate to IssuedDate; null if AppliedDate is missing. Both timestamps are at or before issuance.; totaldaysplanreview: Total plan-review days recorded by SDCI; frequently null in 2016-era rows and left blank.; numberreviewcycles: Number of plan-review correction cycles recorded before issuance; null left blank.; daysoutcorrections: Days the applicant spent outside the city responding to plan-review corrections; missing values left blank.; issue_month_sin: Sine of the issuance month, a cyclical calendar encoding known at issuance.; issue_month_cos: Cosine of the issuance month, a cyclical calendar encoding known at issuance.; originalzip: Five-digit project ZIP mapped to the 20 most frequent training-row ZIPs plus other and missing; the top-20 list is fitted on training rows only.; description_length: Character length of the free-text scope description; null when description is missing.; description_stfi: Binary token for stfi in the scope description; null when description is missing, not zero.; description_per_plan: Binary token for per plan in the scope description; null when description is missing, not zero.; description_apartment: Binary token for apartment in the scope description; null when description is missing, not zero.; description_unit: Binary token for unit in the scope description; null when description is missing, not zero.; description_tenant: Binary token for tenant in the scope description; null when description is missing, not zero.; description_demolition: Binary token for demolition in the scope description; null when description is missing, not zero.; dependentbuilding: Dependent-building indicator recorded at filing; null left blank.; standardplan: Pre-approved DADU standard-plan indicator; null left blank.; parent_permit_flag: One when ParentPermitNum is present, else zero; a filing-structure indicator.; related_mup_flag: One when RelatedMUP is present, else zero; a filing-structure indicator.
Recorded structural checks and baselines
PASSED Gate 1.6.1.
Existing unmodified runner baseline on public validation only; no holdout scoring
These historical checks do not establish availability at each prediction time or protection against every shortcut. Scores below describe the recorded protocol.
Limits
The outcome is an administrative record: a physically finished project whose final inspection was never logged, or whose permit lapsed, counts as non-completion, so the label measures closure of the permit as much as completion of the work.
Current-vintage retrospectivity: EstProjectCost and HousingUnits are documented as subject to change if the project is modified, and the plan-review day columns are frequently null in 2016-era rows. Nothing here proves those fields were visible to an analyst in 2016; only that they are recorded on the current snapshot and are logically prior to issuance.
The dataset is a live, daily-refreshed endpoint with no immutable archive; a rebuild cannot reproduce the byte-identical extract. Labels for eligible cohorts are nonetheless stable because every 48-month window closed before 2026-09-30, so later refreshes can only add completions that fall outside the window.
The snapshot is a rolling window. Pre-2004 rows are sparse and 2004-2015 coverage is uneven, which is why the cohort window starts in 2016; older issuance dates in the file (including 1900-01-01 sentinels) must be filtered, not trusted.
The City's Terms of Use bar commercial use of data that is or can be configured as a list of individuals. Contractor names, applicant-linked fields and street addresses are therefore dropped and only ZIP is kept; this is a preservation obligation on the release, not a formality.
Single city and a single permitting authority, so external validity is limited to Seattle's permit regime; holdout cohorts for 2021-2022 also sit in a different construction-cost and post-pandemic regime than the training years, which is part of the task rather than an artefact to remove.
EstProjectCost >= 50000 is used both for eligibility and as a feature, so the population definition partly depends on a mutable declared field; a small number of projects revised upward after issuance can enter the cohort retroactively.
Deliberate exclusion of cross-row history means contractor track record, site-level phasing and prior-permit outcomes - the strongest real-world predictors - are absent; permits sharing a Development Site or ParentPermitNum remain correlated, so rows are not strictly independent across sibling permits of one project.
ExpiresDate is mechanically IssuedDate plus 18 months in the sampled rows, so permit duration carries almost no signal and is not used as a feature.
A same-day CompletedDate (issued and completed on the same date) occurs in the eligible slice, which suggests some permits are closed administratively on issue; such rows are kept because the documented field rule is unambiguous, but they depress the informativeness of the positive class.
The outcome is an administrative final-inspection completion record, not occupancy or certificate of occupancy. A physically finished project whose final inspection was never logged counts as non-completion.
A null CompletedDate is documented as generally still under inspection; the frozen task maps null to y=0. This is an evidenced field interpretation, not an assumption that an unknown outcome is negative after the 48-month window.
The Socrata resource is a live daily-refreshed current snapshot with no immutable archive. Exact population counts are tied to the 2026-11-14 publisher evidence and may fail on a later refresh.
There is no full-row historical availability ledger. Feature availability at issuance is assumed from field semantics and current-snapshot presence, not proven.
Plan-review and housing-unit fields have substantial missingness. Missing numeric and token values are left blank, not filled with zero.
EstProjectCost is mutable and is used both for population eligibility and as a feature, so a few projects may enter the cohort retrospectively after an upward revision.
Rows may be correlated across sibling permits through ParentPermitNum or RelatedMUP; no site-level independence is claimed.
The 2021-2022 holdout sits in a different construction-cost and post-pandemic regime than the training years. This is part of the task, not an artefact to remove.
ExpiresDate is mechanically IssuedDate plus about 18 months and is not used as a feature.
Same-day CompletedDate values occur; they are kept because the field rule is unambiguous, but they depress the informativeness of the positive class.
No uncertainty interval; observed differences do not establish universal headroom
Source, licences, and citation
City of Seattle (dataset attribution field on the portal; no specific credit is required unless the publishing department indicates otherwise)
Source: Public Domain (Socrata license designation for this dataset; licenseId PUBLIC_DOMAIN), with City of Seattle open-data policy granting publication without registration, license or use restrictions. Compilation: CC0-1.0. Code and documentation: MIT.
Our additional compilation rights only; upstream terms remain in force.