v0.3.0: + Russia (10 districts) + Central America + missing US states/territories (310 units, 16.3M polygons)

.gitattributes

@@ -0,0 +1,8 @@
+combined/all_world.parquet filter=lfs diff=lfs merge=lfs -text
+per_country/bahamas/bahamas.parquet filter=lfs diff=lfs merge=lfs -text
+per_country/belize/belize.parquet filter=lfs diff=lfs merge=lfs -text
+per_country/costa-rica/costa-rica.parquet filter=lfs diff=lfs merge=lfs -text
+per_country/cuba/cuba.parquet filter=lfs diff=lfs merge=lfs -text
+per_country/el-salvador/el-salvador.parquet filter=lfs diff=lfs merge=lfs -text
+per_country/haiti-and-domrep/haiti-and-domrep.parquet filter=lfs diff=lfs merge=lfs -text
+per_country/honduras/honduras.parquet filter=lfs diff=lfs merge=lfs -text

README.md

@@ -19,7 +19,7 @@ size_categories:
 
 # osm-polygon-selection dataset
 
-A curated set of OpenStreetMap polygons from **284
+A curated set of OpenStreetMap polygons from **310
 geographic units** — *sovereign countries plus sub-country regions*
 like Brazilian states, Chinese provinces, Indian zones, US states,
 Canadian provinces, Japanese regions, and Indonesian islands —
@@ -47,9 +47,9 @@ small ponds, single fields) and larger than 100 km² (whole countries,
 mountain ranges, big seas) are excluded by the size filter (see
 [Filter chain](#filter-chain) below).
 
-**Status:** All 284 geographic units are extracted end-to-end.
+**Status:** All 310 geographic units are extracted end-to-end.
 
-**Total polygons:** 14,197,463
+**Total polygons:** 16,297,690
 (combined parquet: `combined/all_world.parquet`).
 
 ## Coverage
@@ -61,15 +61,14 @@ district, island group, zone):
 
 | unit type | count | examples |
 |-----------|------:|----------|
-| Sovereign country | ~180 | france, ghana, japan, peru, australia |
-| Sub-country region | ~100 | brazil-sudeste, china-beijing, japan-kanto, india-central-zone, us-texas, canada-ontario |
-| Multi-country bundle | ~5 | gcc-states, ireland-and-northern-ireland, senegal-and-gambia |
+| Sovereign country | ~170 | france, ghana, japan, peru, australia, brazil, argentina |
+| Sub-country region | ~130 | brazil-sudeste, china-beijing, japan-kanto, india-central-zone, us-texas, canada-ontario, russia-siberian-fed-district |
+| Multi-country bundle | ~10 | gcc-states, ireland-and-northern-ireland, senegal-and-gambia, haiti-and-domrep, malaysia-singapore-brunei, israel-and-palestine |
 
-Total: **284 geographic units** across all 6 inhabited
-continents plus Oceania. The **284** (or whatever current count)
-is **not** the count of sovereign states (which is ~195 per the
-UN); it's the count of discrete Geofabrik PBF regions we
-processed.
+Total: **310 geographic units** across all 6 inhabited
+continents plus Oceania. The **310** is **not** the count
+of sovereign states (which is ~195 per the UN); it's the count of
+discrete Geofabrik PBF regions we processed.
 
 ## Layout
 
@@ -78,10 +77,10 @@ you need:
 
 | folder | what's inside | typical size |
 |--------|---------------|--------------|
-| [`per_country/`](./per_country/) | one folder per country with `<country>.parquet` + `README.md` | ~25 GB total, <1 MB per small country |
-| [`combined/`](./combined/) | `all_world.parquet` — every polygon in one file | ~13 GB |
+| [`per_country/`](./per_country/) | one folder per country with `<country>.parquet` + `README.md` | ~28 GB total, <1 MB per small country |
+| [`combined/`](./combined/) | `all_world.parquet` — every polygon in one file | ~14 GB |
 | [`splits/`](./splits/) | `train.parquet`, `val.parquet`, `test.parquet` — pre-filtered split parquets (no `split` column needed) | ~13 GB total |
-| [`sample/`](./sample/) | `sample_map.jsonl` — ~17k representative polygons for quick viz | ~3 MB |
+| [`sample/`](./sample/) | `sample_map.jsonl` — ~18k representative polygons for quick viz | ~3 MB |
 | [`preview/`](./preview/) | `map_preview.png` — static map thumbnail | ~1 MB |
 
 Start with `sample/` or `preview/` for a quick look. Pull
@@ -117,8 +116,8 @@ reproject it directly without re-deriving from centroid+area.
 ## Provenance
 
 - Pipeline version: v0.1.0
-- Git SHA: d0a04958260fc6edd097ef8c5897e49c1c057150
-- Built: 2026-07-05T17:22:32.399879
+- Git SHA: d69b105c41732c72c2162d737e3353b95bcbdfbf
+- Built: 2026-07-06T23:15:51.406227
 - Source: Geofabrik regional extracts (`https://download.geofabrik.de/`)
 - Whitelist: 22,075 OSM `key=value` tags from osm-stats (see
   `docs/whitelist_decisions.md` in the project repo, or read the
@@ -138,17 +137,17 @@ country. Circle size is proportional to `sqrt(area_km2)`.)
 
 ## Size-bin distribution (full dataset)
 
-Counts every polygon in the **14,197,463-polygon** dataset
+Counts every polygon in the **16,297,690-polygon** dataset
 by `size_bin`, computed directly from `combined/all_world.parquet`
 via `pyarrow.compute.value_counts`. Percentages are exact ratios
 over the entire dataset, not a sample.
 
 | size_bin | count | pct |
 |----------|-------|-----|
-| small | 11,202,879 | 78.9% |
-| medium | 2,572,850 | 18.1% |
-| large | 421,734 | 3.0% |
-| **Total** | 14,197,463 | 100.0% |
+| small | 12,474,300 | 76.5% |
+| medium | 3,294,828 | 20.2% |
+| large | 528,562 | 3.2% |
+| **Total** | 16,297,690 | 100.0% |
 
 
 ## Example row
@@ -198,10 +197,10 @@ one of three values: `train`, `val`, or `test`.
 
 | split | ratio | polygons |
 |-------|-------|----------|
-| train | 80% | 11,358,587 |
-| val   | 10% | 1,419,548 |
-| test  | 10% | 1,419,328 |
-| **Total** | **100%** | **14,197,463** |
+| train | 80% | 13,037,271 |
+| val   | 10% | 1,628,432 |
+| test  | 10% | 1,631,987 |
+| **Total** | **100%** | **16,297,690** |
 
 The split is **stratified by country**: each country's rows are
 assigned to train/val/test independently using a global
@@ -239,9 +238,11 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | austria | 133,711 | clean |
 | azerbaijan | 15,826 | clean |
 | azores | 2,640 | clean |
+| bahamas | 2,882 | clean |
 | bangladesh | 11,444 | clean |
 | belarus | 223,750 | clean |
 | belgium | 125,108 | clean |
+| belize | 12,972 | clean |
 | benin | 4,614 | clean |
 | bhutan | 8,769 | clean |
 | bolivia | 30,701 | clean |
@@ -311,7 +312,9 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | comores | 395 | clean |
 | congo-brazzaville | 6,643 | clean |
 | congo-democratic-republic | 85,106 | clean |
+| costa-rica | 7,476 | clean |
 | croatia | 47,140 | clean |
+| cuba | 23,386 | clean |
 | cyprus | 4,846 | clean |
 | czech-republic | 271,062 | clean |
 | denmark | 175,795 | clean |
@@ -319,6 +322,7 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | east-timor | 1,553 | clean |
 | ecuador | 16,139 | clean |
 | egypt | 24,623 | clean |
+| el-salvador | 3,517 | clean |
 | equatorial-guinea | 1,004 | clean |
 | eritrea | 3,278 | clean |
 | estonia | 47,160 | clean |
@@ -338,6 +342,8 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | guinea | 12,311 | clean |
 | guinea-bissau | 2,109 | clean |
 | guyana | 2,192 | clean |
+| haiti-and-domrep | 4,604 | clean |
+| honduras | 6,160 | clean |
 | hungary | 77,569 | clean |
 | iceland | 47,896 | clean |
 | india-central-zone | 51,347 | clean |
@@ -360,6 +366,7 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | israel-and-palestine | 13,681 | clean |
 | italy | 276,991 | clean |
 | ivory-coast | 14,273 | clean |
+| jamaica | 1,865 | clean |
 | japan-chubu | 14,376 | clean |
 | japan-chugoku | 10,128 | clean |
 | japan-hokkaido | 18,638 | clean |
@@ -408,11 +415,13 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | netherlands | 207,459 | clean |
 | new-caledonia | 2,141 | clean |
 | new-zealand | 127,834 | clean |
+| nicaragua | 10,633 | clean |
 | niger | 14,606 | clean |
 | nigeria | 33,059 | clean |
 | north-korea | 19,295 | clean |
 | norway | 413,801 | clean |
 | pakistan | 36,200 | clean |
+| panama | 6,663 | clean |
 | papua-new-guinea | 9,006 | clean |
 | paraguay | 29,619 | clean |
 | peru | 26,038 | clean |
@@ -421,6 +430,16 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | polynesie-francaise | 2,195 | clean |
 | portugal | 66,287 | clean |
 | romania | 115,401 | clean |
+| russia-central-fed-district | 315,956 | clean |
+| russia-crimean-fed-district | 20,645 | clean |
+| russia-far-eastern-fed-district | 192,337 | clean |
+| russia-kaliningrad | 12,432 | clean |
+| russia-north-caucasus-fed-district | 52,352 | clean |
+| russia-northwestern-fed-district | 381,413 | clean |
+| russia-siberian-fed-district | 291,830 | clean |
+| russia-south-fed-district | 166,644 | clean |
+| russia-ural-fed-district | 184,300 | clean |
+| russia-volga-fed-district | 237,169 | clean |
 | rwanda | 3,976 | clean |
 | saint-helena-ascension-and-tristan-da-cunha | 174 | clean |
 | samoa | 369 | clean |
@@ -458,14 +477,17 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | united-kingdom | 205,002 | clean |
 | uruguay | 10,051 | clean |
 | us-alabama | 28,004 | clean |
+| us-alaska | 62,521 | clean |
 | us-arizona | 38,727 | clean |
 | us-arkansas | 18,513 | clean |
+| us-california | 51,998 | clean |
 | us-colorado | 61,550 | clean |
 | us-connecticut | 12,802 | clean |
 | us-delaware | 7,255 | clean |
 | us-district-of-columbia | 564 | clean |
 | us-florida | 229,464 | clean |
 | us-georgia | 29,945 | clean |
+| us-hawaii | 2,619 | clean |
 | us-idaho | 17,217 | clean |
 | us-illinois | 79,460 | clean |
 | us-indiana | 42,034 | clean |
@@ -491,7 +513,9 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | us-north-dakota | 14,275 | clean |
 | us-ohio | 117,344 | clean |
 | us-oklahoma | 13,752 | clean |
+| us-oregon | 45,342 | clean |
 | us-pennsylvania | 44,038 | clean |
+| us-puerto-rico | 2,318 | clean |
 | us-rhode-island | 6,360 | clean |
 | us-south-carolina | 20,613 | clean |
 | us-south-dakota | 11,975 | clean |
@@ -499,6 +523,7 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | us-texas | 82,221 | clean |
 | us-utah | 22,027 | clean |
 | us-vermont | 7,452 | clean |
+| us-virgin-islands | 193 | clean |
 | us-virginia | 33,844 | clean |
 | us-washington | 51,195 | clean |
 | us-west-virginia | 25,081 | clean |
@@ -511,6 +536,6 @@ uv run python scripts/make_split.py --seed 7   # different reproducible split
 | yemen | 7,008 | clean |
 | zambia | 15,684 | clean |
 | zimbabwe | 7,545 | clean |
-| **Total** | 14,197,463 | |
+| **Total** | 16,297,690 | |
 
 [Back to the dataset root](./README.md)

combined/README.md

@@ -3,8 +3,8 @@
 **1 file**: `all_world.parquet` — a single parquet with **every**
 polygon from **every** country concatenated.
 
-**14,197,463 polygons** across **284 countries and
-sub-country regions** (11.8 GB). Schema is identical to the
+**16,297,690 polygons** across **310 countries and
+sub-country regions** (13.2 GB). Schema is identical to the
 per-country parquets. Built from the per-country parquet files in
 [`../per_country/`](../per_country/). The `country` column tells you
 which country each row came from.

combined/all_world.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:99cbbeac3306a70e1c58c9b93be327075b95ecceaf9010764868c4e3ecf2658c
-size 12679921093
+oid sha256:9a5fd87cec1aa752d0deb9a519a3e7ec4983388e67f601fe86c2896fce0ba4ed
+size 14183595583

manifest.json

@@ -1,9 +1,9 @@
 {
   "version": "v0.1.0",
-  "git_sha": "d0a04958260fc6edd097ef8c5897e49c1c057150",
-  "built_at": "2026-07-05T17:22:32.399879",
-  "total_polygons": 14197463,
-  "n_countries": 284,
+  "git_sha": "d69b105c41732c72c2162d737e3353b95bcbdfbf",
+  "built_at": "2026-07-06T23:15:51.406227",
+  "total_polygons": 16297690,
+  "n_countries": 310,
   "countries": [
     {
       "country": "afghanistan",
@@ -77,6 +77,12 @@
       "extract_status": "clean",
       "pbf_date": "2026-06-26"
     },
+    {
+      "country": "bahamas",
+      "n_polygons": 2882,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "bangladesh",
       "n_polygons": 11444,
@@ -95,6 +101,12 @@
       "extract_status": "clean",
       "pbf_date": "2026-06-26"
     },
+    {
+      "country": "belize",
+      "n_polygons": 12972,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "benin",
       "n_polygons": 4614,
@@ -509,12 +521,24 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-02"
     },
+    {
+      "country": "costa-rica",
+      "n_polygons": 7476,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "croatia",
       "n_polygons": 47140,
       "extract_status": "clean",
       "pbf_date": "2026-06-26"
     },
+    {
+      "country": "cuba",
+      "n_polygons": 23386,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "cyprus",
       "n_polygons": 4846,
@@ -557,6 +581,12 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-02"
     },
+    {
+      "country": "el-salvador",
+      "n_polygons": 3517,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "equatorial-guinea",
       "n_polygons": 1004,
@@ -671,6 +701,18 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-04"
     },
+    {
+      "country": "haiti-and-domrep",
+      "n_polygons": 4604,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
+    {
+      "country": "honduras",
+      "n_polygons": 6160,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "hungary",
       "n_polygons": 77569,
@@ -803,6 +845,12 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-02"
     },
+    {
+      "country": "jamaica",
+      "n_polygons": 1865,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "japan-chubu",
       "n_polygons": 14376,
@@ -1091,6 +1139,12 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-04"
     },
+    {
+      "country": "nicaragua",
+      "n_polygons": 10633,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "niger",
       "n_polygons": 14606,
@@ -1121,6 +1175,12 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-04"
     },
+    {
+      "country": "panama",
+      "n_polygons": 6663,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "papua-new-guinea",
       "n_polygons": 9006,
@@ -1169,6 +1229,66 @@
       "extract_status": "clean",
       "pbf_date": "2026-06-26"
     },
+    {
+      "country": "russia-central-fed-district",
+      "n_polygons": 315956,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
+    {
+      "country": "russia-crimean-fed-district",
+      "n_polygons": 20645,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-05"
+    },
+    {
+      "country": "russia-far-eastern-fed-district",
+      "n_polygons": 192337,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-05"
+    },
+    {
+      "country": "russia-kaliningrad",
+      "n_polygons": 12432,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-05"
+    },
+    {
+      "country": "russia-north-caucasus-fed-district",
+      "n_polygons": 52352,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-05"
+    },
+    {
+      "country": "russia-northwestern-fed-district",
+      "n_polygons": 381413,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-05"
+    },
+    {
+      "country": "russia-siberian-fed-district",
+      "n_polygons": 291830,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-05"
+    },
+    {
+      "country": "russia-south-fed-district",
+      "n_polygons": 166644,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-05"
+    },
+    {
+      "country": "russia-ural-fed-district",
+      "n_polygons": 184300,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-05"
+    },
+    {
+      "country": "russia-volga-fed-district",
+      "n_polygons": 237169,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-05"
+    },
     {
       "country": "rwanda",
       "n_polygons": 3976,
@@ -1391,6 +1511,12 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-05"
     },
+    {
+      "country": "us-alaska",
+      "n_polygons": 62521,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "us-arizona",
       "n_polygons": 38727,
@@ -1403,6 +1529,12 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-05"
     },
+    {
+      "country": "us-california",
+      "n_polygons": 51998,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "us-colorado",
       "n_polygons": 61550,
@@ -1439,6 +1571,12 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-05"
     },
+    {
+      "country": "us-hawaii",
+      "n_polygons": 2619,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "us-idaho",
       "n_polygons": 17217,
@@ -1589,12 +1727,24 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-05"
     },
+    {
+      "country": "us-oregon",
+      "n_polygons": 45342,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "us-pennsylvania",
       "n_polygons": 44038,
       "extract_status": "clean",
       "pbf_date": "2026-07-05"
     },
+    {
+      "country": "us-puerto-rico",
+      "n_polygons": 2318,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "us-rhode-island",
       "n_polygons": 6360,
@@ -1637,6 +1787,12 @@
       "extract_status": "clean",
       "pbf_date": "2026-07-05"
     },
+    {
+      "country": "us-virgin-islands",
+      "n_polygons": 193,
+      "extract_status": "clean",
+      "pbf_date": "2026-07-06"
+    },
     {
       "country": "us-virginia",
       "n_polygons": 33844,

per_country/README.md

@@ -1,8 +1,8 @@
 # per_country/
 
-**284 country folders** (one per country),
+**310 country folders** (one per country),
 each containing one `<country>.parquet` file and one `README.md`.
-Total polygons across all countries: **14,197,463**.
+Total polygons across all countries: **16,297,690**.
 
 The per-country split mirrors Geofabrik's regional extracts.
 Use this folder when you want a single country without paying for
@@ -11,7 +11,7 @@ Each parquet has the same 13-column schema as
 [`../combined/all_world.parquet`](../combined/all_world.parquet).
 
 For the all-in-one file, see [`../combined/`](../combined/). For a
-small representative sample (16,734 polygons, one per
+small representative sample (17,911 polygons, one per
 ~1.7k), see [`../sample/`](../sample/). For a thumbnail of the
 geographic distribution, see [`../preview/`](../preview/).
 

per_country/bahamas/README.md

@@ -0,0 +1,36 @@
+# bahamas
+
+2,882 polygons from `bahamas-latest.osm.pbf` on Geofabrik
+([source](https://download.geofabrik.de/central-america/bahamas.html)), filtered by the
+[22,075-tag whitelist](https://github.com/NoeFlandre/osm-stats) and
+classified by continent + size bin.
+
+| field | value |
+|-------|-------|
+| country | `bahamas` |
+| polygons | **2,882** |
+| extract_status | **clean** |
+| pbf_date | 2026-07-06 |
+| source | `bahamas-latest.osm.pbf` |
+| Geofabrik extract | <https://download.geofabrik.de/central-america/bahamas.html> |
+
+## Geometry
+
+The parquet file in this folder
+(`bahamas.parquet`) has the same schema as the combined
+`combined/all_world.parquet`. Each row carries the polygon
+**geometry as WKT** (default), plus centroid + area + the
+whitelist-matched tag.
+
+Load with:
+
+```python
+import pyarrow.parquet as pq
+table = pq.read_table("per_country/bahamas/bahamas.parquet")
+df = table.to_pandas()
+# df["geometry_wkt"] is a column of WKT strings; parse with shapely.wkt.loads
+```
+
+## Notes
+
+Bahamas has 2,882 polygons in this dataset. Extract status: **clean**. Source: Geofabrik [`bahamas-latest.osm.pbf`](https://download.geofabrik.de/central-america/bahamas.html).

per_country/bahamas/bahamas.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4cb1523200db77d06e8b42faf19826dec1fd871634e643c6dfa7f387f206d6d
+size 3771732

per_country/belize/README.md

@@ -0,0 +1,36 @@
+# belize
+
+12,972 polygons from `belize-latest.osm.pbf` on Geofabrik
+([source](https://download.geofabrik.de/central-america/belize.html)), filtered by the
+[22,075-tag whitelist](https://github.com/NoeFlandre/osm-stats) and
+classified by continent + size bin.
+
+| field | value |
+|-------|-------|
+| country | `belize` |
+| polygons | **12,972** |
+| extract_status | **clean** |
+| pbf_date | 2026-07-06 |
+| source | `belize-latest.osm.pbf` |
+| Geofabrik extract | <https://download.geofabrik.de/central-america/belize.html> |
+
+## Geometry
+
+The parquet file in this folder
+(`belize.parquet`) has the same schema as the combined
+`combined/all_world.parquet`. Each row carries the polygon
+**geometry as WKT** (default), plus centroid + area + the
+whitelist-matched tag.
+
+Load with:
+
+```python
+import pyarrow.parquet as pq
+table = pq.read_table("per_country/belize/belize.parquet")
+df = table.to_pandas()
+# df["geometry_wkt"] is a column of WKT strings; parse with shapely.wkt.loads
+```
+
+## Notes
+
+Belize has 12,972 polygons in this dataset. Extract status: **clean**. Source: Geofabrik [`belize-latest.osm.pbf`](https://download.geofabrik.de/central-america/belize.html).

per_country/belize/belize.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9921160c3e63f6af8e68ec4383b08b516e178732ef9487f36646ef7620940210
+size 13225653

per_country/costa-rica/README.md

@@ -0,0 +1,36 @@
+# costa-rica
+
+7,476 polygons from `costa-rica-latest.osm.pbf` on Geofabrik
+([source](https://download.geofabrik.de/central-america/costa-rica.html)), filtered by the
+[22,075-tag whitelist](https://github.com/NoeFlandre/osm-stats) and
+classified by continent + size bin.
+
+| field | value |
+|-------|-------|
+| country | `costa-rica` |
+| polygons | **7,476** |
+| extract_status | **clean** |
+| pbf_date | 2026-07-06 |
+| source | `costa-rica-latest.osm.pbf` |
+| Geofabrik extract | <https://download.geofabrik.de/central-america/costa-rica.html> |
+
+## Geometry
+
+The parquet file in this folder
+(`costa-rica.parquet`) has the same schema as the combined
+`combined/all_world.parquet`. Each row carries the polygon
+**geometry as WKT** (default), plus centroid + area + the
+whitelist-matched tag.
+
+Load with:
+
+```python
+import pyarrow.parquet as pq
+table = pq.read_table("per_country/costa-rica/costa-rica.parquet")
+df = table.to_pandas()
+# df["geometry_wkt"] is a column of WKT strings; parse with shapely.wkt.loads
+```
+
+## Notes
+
+Costa-Rica has 7,476 polygons in this dataset. Extract status: **clean**. Source: Geofabrik [`costa-rica-latest.osm.pbf`](https://download.geofabrik.de/central-america/costa-rica.html).

per_country/costa-rica/costa-rica.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3330e765b3898ede3b5865d2cf7485565422635281d948fd1492b32103f02c7b
+size 8443099

per_country/cuba/README.md

@@ -0,0 +1,36 @@
+# cuba
+
+23,386 polygons from `cuba-latest.osm.pbf` on Geofabrik
+([source](https://download.geofabrik.de/central-america/cuba.html)), filtered by the
+[22,075-tag whitelist](https://github.com/NoeFlandre/osm-stats) and
+classified by continent + size bin.
+
+| field | value |
+|-------|-------|
+| country | `cuba` |
+| polygons | **23,386** |
+| extract_status | **clean** |
+| pbf_date | 2026-07-06 |
+| source | `cuba-latest.osm.pbf` |
+| Geofabrik extract | <https://download.geofabrik.de/central-america/cuba.html> |
+
+## Geometry
+
+The parquet file in this folder
+(`cuba.parquet`) has the same schema as the combined
+`combined/all_world.parquet`. Each row carries the polygon
+**geometry as WKT** (default), plus centroid + area + the
+whitelist-matched tag.
+
+Load with:
+
+```python
+import pyarrow.parquet as pq
+table = pq.read_table("per_country/cuba/cuba.parquet")
+df = table.to_pandas()
+# df["geometry_wkt"] is a column of WKT strings; parse with shapely.wkt.loads
+```
+
+## Notes
+
+Cuba has 23,386 polygons in this dataset. Extract status: **clean**. Source: Geofabrik [`cuba-latest.osm.pbf`](https://download.geofabrik.de/central-america/cuba.html).

per_country/cuba/cuba.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9b998ad5bfefab88f00783ae9abe88866782919159e6a9bc54eb141d0f7cc36
+size 14823404

per_country/el-salvador/README.md

@@ -0,0 +1,36 @@
+# el-salvador
+
+3,517 polygons from `el-salvador-latest.osm.pbf` on Geofabrik
+([source](https://download.geofabrik.de/central-america/el-salvador.html)), filtered by the
+[22,075-tag whitelist](https://github.com/NoeFlandre/osm-stats) and
+classified by continent + size bin.
+
+| field | value |
+|-------|-------|
+| country | `el-salvador` |
+| polygons | **3,517** |
+| extract_status | **clean** |
+| pbf_date | 2026-07-06 |
+| source | `el-salvador-latest.osm.pbf` |
+| Geofabrik extract | <https://download.geofabrik.de/central-america/el-salvador.html> |
+
+## Geometry
+
+The parquet file in this folder
+(`el-salvador.parquet`) has the same schema as the combined
+`combined/all_world.parquet`. Each row carries the polygon
+**geometry as WKT** (default), plus centroid + area + the
+whitelist-matched tag.
+
+Load with:
+
+```python
+import pyarrow.parquet as pq
+table = pq.read_table("per_country/el-salvador/el-salvador.parquet")
+df = table.to_pandas()
+# df["geometry_wkt"] is a column of WKT strings; parse with shapely.wkt.loads
+```
+
+## Notes
+
+El-Salvador has 3,517 polygons in this dataset. Extract status: **clean**. Source: Geofabrik [`el-salvador-latest.osm.pbf`](https://download.geofabrik.de/central-america/el-salvador.html).

per_country/el-salvador/el-salvador.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af1131a8f619f652c21a103dd0380e731df9aefca505c7dc2420bd68079b6e7b
+size 7311854

per_country/haiti-and-domrep/README.md

@@ -0,0 +1,36 @@
+# haiti-and-domrep
+
+4,604 polygons from `haiti-and-domrep-latest.osm.pbf` on Geofabrik
+([source](https://download.geofabrik.de/central-america/haiti-and-domrep.html)), filtered by the
+[22,075-tag whitelist](https://github.com/NoeFlandre/osm-stats) and
+classified by continent + size bin.
+
+| field | value |
+|-------|-------|
+| country | `haiti-and-domrep` |
+| polygons | **4,604** |
+| extract_status | **clean** |
+| pbf_date | 2026-07-06 |
+| source | `haiti-and-domrep-latest.osm.pbf` |
+| Geofabrik extract | <https://download.geofabrik.de/central-america/haiti-and-domrep.html> |
+
+## Geometry
+
+The parquet file in this folder
+(`haiti-and-domrep.parquet`) has the same schema as the combined
+`combined/all_world.parquet`. Each row carries the polygon
+**geometry as WKT** (default), plus centroid + area + the
+whitelist-matched tag.
+
+Load with:
+
+```python
+import pyarrow.parquet as pq
+table = pq.read_table("per_country/haiti-and-domrep/haiti-and-domrep.parquet")
+df = table.to_pandas()
+# df["geometry_wkt"] is a column of WKT strings; parse with shapely.wkt.loads
+```
+
+## Notes
+
+Haiti-And-Domrep has 4,604 polygons in this dataset. Extract status: **clean**. Source: Geofabrik [`haiti-and-domrep-latest.osm.pbf`](https://download.geofabrik.de/central-america/haiti-and-domrep.html).

per_country/haiti-and-domrep/haiti-and-domrep.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43e8a11623f5445f821540202eafbee72ce7090636a39083dc08e6ccf581a4ce
+size 3502652

per_country/honduras/README.md

@@ -0,0 +1,36 @@
+# honduras
+
+6,160 polygons from `honduras-latest.osm.pbf` on Geofabrik
+([source](https://download.geofabrik.de/central-america/honduras.html)), filtered by the
+[22,075-tag whitelist](https://github.com/NoeFlandre/osm-stats) and
+classified by continent + size bin.
+
+| field | value |
+|-------|-------|
+| country | `honduras` |
+| polygons | **6,160** |
+| extract_status | **clean** |
+| pbf_date | 2026-07-06 |
+| source | `honduras-latest.osm.pbf` |
+| Geofabrik extract | <https://download.geofabrik.de/central-america/honduras.html> |
+
+## Geometry
+
+The parquet file in this folder
+(`honduras.parquet`) has the same schema as the combined
+`combined/all_world.parquet`. Each row carries the polygon
+**geometry as WKT** (default), plus centroid + area + the
+whitelist-matched tag.
+
+Load with:
+
+```python
+import pyarrow.parquet as pq
+table = pq.read_table("per_country/honduras/honduras.parquet")
+df = table.to_pandas()
+# df["geometry_wkt"] is a column of WKT strings; parse with shapely.wkt.loads
+```
+
+## Notes
+
+Honduras has 6,160 polygons in this dataset. Extract status: **clean**. Source: Geofabrik [`honduras-latest.osm.pbf`](https://download.geofabrik.de/central-america/honduras.html).

per_country/honduras/honduras.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f73bc9165a7765ab565b9830be2a68b793452380ff2ff0517d69f1a90029db99
+size 4405784