HLS4ML LHC Jet Dataset: High-pT Jets with 150 Particles per Jet

Simulated proton-proton collisions from the Large Hadron Collider, prepared for FastML/HLS4ML studies. The dataset includes high-level features, jet images, and particle lists for high-pT jets. It was created by M. Pierini of the European Organization for Nuclear Research.

Use Cases

• Training neural networks for jet classification based on high-level features.
• Developing image-based models for particle physics using jet images.
• Benchmarking hardware-accelerated inference models with particle list data.
• Studying jet substructure and properties using the list of up to 150 particles per jet.

Strengths

• Includes three distinct data representations: high-level features, images, and particle lists.
• Contains data for high-pT jets, a key signature in particle physics.
• Specifically prepared for FastML/HLS4ML benchmarking studies.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

European Organization for Nuclear Research (CERN)

Collection Method

Simulations of LHC proton-proton collisions.

Time Range

null

Freshness

Last updated date is unknown.

Geography

null