THULab/rescast_100k_refit

REFIT (TsFile)

Apache TsFile version of Jainam03/RESCAST-100k-REFIT, one of the processed real residential building-energy corpora used with RESCAST-100k.

Overview

Processed real residential time-series subset used with RESCAST-100k. It is derived from the REFIT smart-home electricity dataset.

• Buildings: 20
• Rows: 1,044,201
• Resolution: 15 minutes
• Time span: Per-building ranges vary (roughly 2013-2015); 2013-10-09 13:00 to 2015-07-10 11:45 in the sampled file.
• TsFile layout: The converted corpus is stored as 1 TsFile shard(s) (rescast_100k_refit_000.tsfile), all under the TsFile table rescast_100k_refit.

Each row is one timestamped reading for one building. The TsFile table name is rescast_100k_refit and the building_id TAG identifies the building.

Schema (TsFile structure)

• Time (INT64, milliseconds) — timestamp parsed from the source Time column (UTC for timezone-aware sources).

• building_id (TAG) — building identifier. For the per-building repositories it is the numeric N from the source file name <prefix>_home_N.parquet; for the sharded sample it is the source building_id column.

• Measurements (FIELD):

• total_load (DOUBLE) — total electricity load, kWh

Query one building with a predicate such as WHERE building_id = 1. The source column names contain spaces (and : for the sharded sample); they are written as TsFile-safe identifiers by replacing runs of non-alphanumeric characters with _ and lower-casing, e.g. Indoor Temp -> indoor_temp and Fuel Use: Electricity: Total -> fuel_use_electricity_total.

Static building features

The source repositories also contain house_features_*.parquet (static per-building metadata such as weather location, geometry, HVAC type and insulation). That table is not a time series and is not included in this TsFile repository; it remains in the original dataset.

Usage

Install the Apache TsFile Python SDK (pip install tsfile) and read a converted file:

from pathlib import Path
from tsfile import TsFileReader

path = Path("rescast_100k_refit_000.tsfile")
with TsFileReader(str(path)) as reader:
    schemas = reader.get_all_table_schemas()
    print("tables:", list(schemas))
    table_name = next(iter(schemas))
    table = schemas[table_name]
    columns = [column.get_column_name() for column in table.get_columns()]
    print("columns:", columns)
    field_names = [
        column.get_column_name()
        for column in table.get_columns()
        if column.get_column_name() not in {"Time", "time"}
    ]
    if field_names:
        with reader.query_table(table_name, field_names[:3], batch_size=1024) as result:
            batch = result.read_arrow_batch()
            if batch is not None:
                print(batch.to_pandas().head())

Source & license

• Original dataset: Jainam03/RESCAST-100k-REFIT
• Author / publisher: Murray, D.; Stankovic, L.; Stankovic, V.; processed release by Jainam03
• Paper: Murray et al. (2017). An electrical load measurements dataset of United Kingdom households from a two-year longitudinal study. Scientific Data 4:160122. https://doi.org/10.1038/sdata.2016.122
• License: cc-by-4.0 (processed release; check the original source terms before redistribution)