Datasets:

Modalities:
Image
Text
Formats:
json
DOI:
Libraries:
Datasets
pandas
License:
timeguessr / README.md
titouv's picture
Create README.md
14734a5 verified
metadata
license: apache-2.0

Timeguessr Scraped Dataset

File: dataset.jsonl
Format: JSON Lines – one JSON object per line.

Overview

A collection of images with metadata for the Timeguessr game. Each entry contains a URL to the image, the year it was taken, geographic coordinates, a short description, licensing info, and the country.

Schema

Field Type Description
URL string Direct link to the image (publicly accessible).
Year string Four‑digit year (e.g., "1990").
Location object { "lat": <float>, "lng": <float> } (WGS‑84).
Description string Human‑readable caption.
License string Attribution text supplied by the submitter.
Country string ISO‑3166‑1 alpha‑3 or common name (e.g., "USA").

All fields are required; missing values break downstream pipelines.

Usage

import json

def load_dataset(path):
    with open(path, "r", encoding="utf-8") as f:
        for line in f:
            yield json.loads(line)

# Example: count entries per country
from collections import Counter
counter = Counter()
for record in load_dataset("dataset.jsonl"):
    counter[record["Country"]] += 1
print(counter.most_common())