Test Dataset for MRG-GCN: Urban Cell Graphs from Beijing and London

447.8 MB of test data in PT and PKL formats for evaluating the MRG-GCN model. The dataset provides examples for testing interpretable graph neural networks that reveal multiscale organization in urban systems. It was authored by Xinzhuo Zhao and last updated on figshare in April 2026.

Use Cases

• Testing the MRG-GCN model's ability to reveal multiscale organization in urban systems.
• Benchmarking graph neural network performance on urban data from distinct cities.
• Developing interpretable AI methods for urban planning and systems analysis.

Strengths

• Dataset size is 447.8 MB, indicating a substantial test set.
• Includes data from two distinct urban cells (Beijing and London) for comparative testing.
• Released under a permissive CC-BY-4.0 license.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Likely created as synthetic or processed test data for a specific model.

Time Range

null

Freshness

Last updated 2026-04-15 06:57:09; freshness should be verified.

Geography

Includes urban cells from Beijing and London.