Overview
MassSpecGym provides MS/MS data and evaluation infrastructure for molecule generation, molecule retrieval and spectrum simulation.
Python dataset loaders, transforms and PyTorch Lightning data modules support predefined benchmark tasks; a separate leaderboard presents submitted results.
Pin dataset revision, split and task metrics. The current dataset card describes a v1.5 correction; comparisons across revisions need care. Code and dataset card declare MIT; companion model weights require separate checks.
Key Features
- Python dataset loaders, transforms and PyTorch Lightning data modules support predefined benchmark tasks; a separate leaderboard presents submitted results.
Use Cases
- Compare MS/MS learning methods under a fixed benchmark protocol.
How to Use
Follow the documented Python 3.11 environment and load the chosen dataset revision. Select a task and preserve its train, validation and test split.
See the official repository for installation and interface details: https://github.com/pluskal-lab/MassSpecGym
Sources and verification
This entry describes capabilities documented by the linked project sources. Catalog edits are not evidence of a successful run, compatibility test or scientific performance evaluation.
Read the editorial and verification policy