Road Safety

#	year	topic	category	value
loading...

COLUMN	TYPE	DESCRIPTION	UNIT
year	int64	Calendar year (2023-2024).	-
topic	string	Values: accidents (by severity), deaths (deaths and fatality rate), fire (fire accidents).	-
category	string	Severity/metric within the topic (e.g. fatal, serious_injury, total_death, death_rate, fire_incidents).	-
value	float64	Count for the given year, topic, and category. The death_rate category is a fatality rate per 10,000, not a count.	-

import imibare as imi

df = imi.load("rw.nisr.road-safety.annual")
print(df.head())

import duckdb

duckdb.sql("""
    ATTACH 'https://catalog.cloudflarestorage.com/imibare-data'
    AS imi (TYPE ICEBERG, READ_ONLY)
""")

duckdb.sql("SELECT * FROM imi.rw_nisr_road-safety_annual LIMIT 10").show()

curl -L "https://api.imibare.org/v1/data/rw.nisr.road-safety.annual/download" \
  -o "rw.nisr.road-safety.annual.parquet"

# List all datasets
curl "https://api.imibare.org/v1/datasets"

# Get metadata for this dataset
curl "https://api.imibare.org/v1/datasets/rw.nisr.road-safety.annual"

Every version is a change in the data, and every one is retained; snapshots are never overwritten. Each entry is a dated Parquet snapshot in the dataset's versions/ archive.

2026-07-03 Jul 3, 2026 LATEST

20 rows · 3.0 KB · initial version