KOLCO CLUB · Spatial reasoning

300,000 floor plans.On real CAD layers.

Indian residential plans, studio to four bedroom, drawn as editable SVG with every room name, dimension, area figure and schedule row as live text — and every wall, door, column and fixture on its own standard architectural layer.

300,000editable SVG
floor plans
237–2,186sq ft carpet area
median 1,123
20CAD layers on
every plan
42 GBwith geometry
and metadata
§ 01Structure

Not a picture of a plan. A plan.

Auto-traced floor plans give you one undifferentiated soup of paths. These are drawn on the AIA layer convention, so walls are separable from furniture, and annotation is separable from architecture, without segmenting anything first. All 50 plans in the public sample carry 20 of these layers; none carries fewer.

A-WALLWall geometry
A-WALL-FILLWall poche
A-DOORDoors and swings
A-GLAZWindows and glazing
S-COLSStructural columns
A-FURNFurniture
A-FLOR-FIXTFixtures
A-FLOR-WETWet areas
A-FLOR-BALCBalconies
A-FLOR-BALC-RAILBalcony railings
A-FLOR-COVRFloor coverings
A-FLOR-PLATPlatforms
A-ANNO-DIMSDimension chains
A-ANNO-TEXTRoom labels
A-ANNO-SCHEDDoor and window schedules
A-ANNO-NORTHNorth arrow
A-ANNO-SCALEBARScale bar
A-ANNO-OPENING-MARKSOpening marks

Room labels come as a regular quartet — name, metric dimensions, imperial dimensions, area — so room polygons, adjacency graphs and area regression targets are all recoverable from the drawing itself. Median 139 live text nodes and 275 path elements per plan, measured on the sample.

§ 02Distribution

What 300,000 plans actually look like.

Stated up front, because you will compute it in your first evaluation hour. The distribution follows what actually gets built in India, which makes it realistic and unbalanced at the same time.

By unit type

3 BHK153,454
2 BHK72,372
4 BHK54,955
1 BHK15,315
Studio3,904

By footprint shape

L-shaped209,879
T-shaped65,694
Rectangular17,600
Stepped4,015
U-shaped1,485
Notched1,327

By carpet area

under 50010,323
500-7498,989
750-99937,057
1,000-1,249167,452
1,250-1,49950,226
1,500-1,74923,268
1,750+2,685

Three-bedroom plans and L-shaped footprints dominate, and areas cluster hard between 1,000 and 1,249 sq ft. If your model needs a flat distribution rather than a realistic one, a release can be sampled to a target balance, or the thin classes drawn to order.

§ 03Metadata

Twenty-one fields per plan.

One CSV row per file, shipped with the pack. This is a real row, not a schema sketch.

  • Programme Unit type, rooms, toilets, attached toilets, balconies
  • Area Carpet area in square feet and square metres
  • Geometry Footprint shape, circulation type
  • Orientation Facing, vastu alignment
  • Drawing Dimension system, palette, typeface, symbol style, detail level
  • File Score and size in KB
§ 04Uses

What it is for.

01

Plan understanding

Room detection, area regression, unit-type classification and adjacency extraction, with labels that come from the drawing rather than from a human annotator’s guess.

02

Plan generation

Conditional layout generation — given a footprint, an area budget and a room programme, produce a plan. The metadata gives you the conditioning variables directly.

03

Raster to CAD

Every plan renders deterministically to an image, so each file is an aligned pair for training image-to-vector conversion on technical drawings rather than on logos.

04

Document and drawing VQA

Questions with checkable answers — how many bedrooms, which way does the master face, what is the balcony area — grounded in live text you can read without OCR.