---
id: "house-sales"
slug: "house-sales"
name: "House Sales"
domain: "Real Estate & Property Valuation"
taskType: "regression"
numRows: 21613
numColumns: 21
targetColumn: "sale_price_usd"
primaryMetric: "RMSE"
primaryScore: "111,524"
parquetUrl: "https://huggingface.co/datasets/carlahq/demo-tabular-benchmarks/resolve/main/data/house-sales/house-sales.parquet"
spreadsheetUrl: "https://docs.google.com/spreadsheets/d/1YoZidTrfei6D03y-GXN2zVEZUVqJhfbQMRK2TfRWO5Q/edit?usp=sharing"
canonicalUrl: "https://carlahq.eu/datasets/house-sales"
---

# Dataset: House Sales (house-sales)

> The King County House Sales dataset provides transaction records for 21,613 residential properties sold between May 2014 and May 2015 in Greater Seattle. It serves as a gold-standard benchmark for regression models predicting property market values from structural, geospatial, and neighborhood features.

## Dataset Overview

| Attribute | Value |
| :--- | :--- |
| **Task Type** | `regression` |
| **Business Domain** | Real Estate & Property Valuation |
| **Target Column** | `sale_price_usd` |
| **Rows (Instances)** | 21,613 |
| **Features (Columns)** | 21 |
| **Primary Evaluation Metric** | RMSE |
| **Best Benchmark Score** | **111,524** |

## Business Decision Context
**Context:** Automated Valuation Models (AVMs) are central to modern mortgage underwriting, real estate brokerage, iBuying platforms, and property tax assessments. In high-velocity urban markets like King County / Seattle, manual property appraisal is cost-prohibitive and slow, creating substantial demand for automated, highly accurate continuous price estimation.
**Objective:** Predict the final residential property sale price (in USD) using 19 structural, temporal, geospatial, and neighborhood characteristics.
**Decision Impact:** Underpricing leads to lost commissions and seller margin erosion, whereas overpricing extends days on market (DOM) and creates mortgage default and liquidity exposure for institutional portfolio buyers. High-accuracy AVMs reduce appraisal turnaround from weeks to seconds and minimize variance in automated loan-to-value (LTV) calculations.

## Feature Schema
| Column Name | Data Type | Role | Distinct Values | Missing Values (%) | Description |
| :--- | :--- | :--- | :--- | :--- | :--- |
| `sale_price_usd` | `numeric` | target | 4,028 | 0 | Final realized sale transaction price of the home in US Dollars. |
| `property_id` | `numeric` | feature | 21,436 | 0 | Unique administrative parcel identifier assigned to each residential property in King County. |
| `sale_date` | `string` | feature | 372 | 0 | Timestamp indicating when the property transaction occurred (format: YYYYMMDDT000000). |
| `bedroom_count` | `numeric` | feature | 13 | 0 | Total count of dedicated bedrooms in the house. |
| `bathroom_count` | `numeric` | feature | 30 | 0 | Total number of bathrooms, where 0.5 represents a toilet and sink (half bath) and 0.75 represents a shower without a tub. |
| `living_area_sqft` | `numeric` | feature | 1,038 | 0 | Total interior living space measurement of the home. |
| `lot_area_sqft` | `numeric` | feature | 9,782 | 0 | Total surface area of the land parcel associated with the property. |
| `floor_count` | `numeric` | feature | 6 | 0 | Total levels or stories in the residential building. |
| `is_waterfront` | `numeric` | feature | 2 | 0 | Binary indicator representing whether the property borders a waterfront (1 = Yes, 0 = No). |
| `view_rating` | `numeric` | feature | 5 | 0 | Qualitative index from 0 to 4 grading how scenic the property's surrounding view is. |
| `condition_rating` | `numeric` | feature | 5 | 0 | Physical state assessment of the property on an ordinal scale from 1 (poor/uninhabitable) to 5 (excellent). |
| `construction_grade` | `numeric` | feature | 12 | 0 | Official King County grading scale reflecting construction materials, craftsmanship, and architectural design quality (1-3 poor, 7 average, 11-13 luxury/mansion). |
| `above_ground_sqft` | `numeric` | feature | 946 | 0 | Square footage of interior living housing space situated entirely above ground level. |
| `basement_sqft` | `numeric` | feature | 306 | 0 | Square footage of interior living or storage housing space situated below ground level. |
| `year_built` | `numeric` | feature | 116 | 0 | The calendar year in which the primary building construction was completed. |
| `year_renovated` | `numeric` | feature | 70 | 0 | The calendar year of the most recent major structural restoration or remodel (0 if never renovated). |
| `postal_code` | `numeric` | feature | 70 | 0 | United States Postal Service 5-digit postal routing code representing the local micro-neighborhood. |
| `latitude` | `numeric` | feature | 5,034 | 0 | Geographic coordinate indicating the north-south position of the parcel. |
| `longitude` | `numeric` | feature | 752 | 0 | Geographic coordinate indicating the east-west position of the parcel. |
| `nearest_15_living_sqft` | `numeric` | feature | 777 | 0 | Average interior living square footage of the 15 closest neighboring residential properties. |
| `nearest_15_lot_sqft` | `numeric` | feature | 8,689 | 0 | Average land lot square footage of the 15 closest neighboring residential properties. |

## Empirical Benchmark Leaderboard
| Rank | Algorithm / Model | Model Family | Runtime | R² Score | MAE | RMSE |
| :--- | :--- | :--- | :--- | :--- | :--- | :--- |
| 1 | **Google TabFM v1.0** | Tabular Foundation Model | PyTorch/Python | **0.9273** | 54,970.9 | 98,886.8 |
| 2 | **EXAONE Tabular** | Tabular Foundation Model | PyTorch/Python | **0.9264** | 55,303.3 | 99,519.6 |
| 3 | **TabPFN v3** | Tabular Foundation Model | PyTorch/Python | **0.9225** | 56,817 | 102,039.4 |
| 4 | **nanotabicl Vanilla** | Tabular Foundation Model | PyTorch/Python | **0.9078** | 58,724.6 | 111,387 |
| 5 | **Full TabICLv2 (PyTorch Reference)** | Tabular Foundation Model | PyTorch/Python | **0.9076** | 58,666.4 | 111,523.7 |
| 6 | **Streaming nanotabicl** | Tabular Foundation Model | PyTorch/Python | **0.9017** | 59,637.7 | 115,032.2 |
| 7 | **Carla Engine (TabICLv2 WebGPU)** | Tabular Foundation Model | Browser | **0.9002** | 60,293.5 | 115,866.2 |

### Repeated Cross-Validation Fold Scores (TabICLv2 Reference)
| Fold | Train Samples | Test Samples | Metric Score | Latency |
| :--- | :--- | :--- | :--- | :--- |
| Fold 0 | 14408 | 7205 | 109300.4366 | 7407ms |
| Fold 1 | 14409 | 7204 | 124408.0531 | 7468ms |
| Fold 2 | 14409 | 7204 | 99937.1878 | 7499ms |
| Fold 0 | 14408 | 7205 | 102848.0776 | 7458ms |
| Fold 1 | 14409 | 7204 | 112711.4560 | 7530ms |
| Fold 2 | 14409 | 7204 | 121621.2060 | 7522ms |
| Fold 0 | 14408 | 7205 | 111724.3222 | 7631ms |
| Fold 1 | 14409 | 7204 | 105725.6564 | 7571ms |
| Fold 2 | 14409 | 7204 | 115437.1858 | 7632ms |

## Provenance & Downloadable Artifacts
- **Dataset Parquet File:** [Direct Parquet Download](https://huggingface.co/datasets/carlahq/demo-tabular-benchmarks/resolve/main/data/house-sales/house-sales.parquet)
- **Google Sheets Interactive Demo:** [Open in Google Sheets](https://docs.google.com/spreadsheets/d/1YoZidTrfei6D03y-GXN2zVEZUVqJhfbQMRK2TfRWO5Q/edit?usp=sharing)
- **OpenML Source:** [OpenML Dataset](https://www.openml.org/d/42092)
- **Academic Paper / Citation:** [Read Research Paper](https://www.openml.org/d/42092)
- **Canonical HTML Page:** [https://carlahq.eu/datasets/house-sales](https://carlahq.eu/datasets/house-sales)