datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
tokamark-v1
TokaMark
TokaMark is a large, open benchmark dataset for evaluating AI models on real experimental diagnostics from the Mega Ampere Spherical Tokamak (MAST). It packages 39 harmonized signals from 11,573 MAST shots into a standardized, task-ready format spanning 14 downstream tasks across plasma equilibrium reconstruction, short-term magnetics dynamics, kinetic profile evolution, and long-range forecasting of MHD activity and disruptions.
This repository hosts the data release… See the full description on the dataset page: https://huggingface.co/datasets/UKAEA-IBM-STFC/tokamark-v1.tokamark-robustness-data
TokaMark Sensor Robustness Benchmark Data
Associated paper: Benchmarking Sensor Robustness in Plasma Diagnostic Models: A Systematic Evaluation on TokaMarkAuthor: Neerav GuptaCode: github.com/Neerav-Gupta/tokamark-robustness
Dataset Description
This dataset contains pre-processed numpy arrays, trained model checkpoints, and experiment results from the first systematic robustness benchmark of plasma diagnostic ML models under realistic sensor failure, using the… See the full description on the dataset page: https://huggingface.co/datasets/Neerav-Gupta/tokamark-robustness-data.tokamark-dataset
TokaMark: A Comprehensive Benchmark for MAST Tokamak Plasma Models
Abstract: Development and operation of commercially viable fusion energy reactors such as tokamaks require accurate predictions of plasma dynamics from sparse, noisy, and incomplete sensors readings. The complexity of the underlying physics and the heterogeneity of experimental data pose formidable challenges for conventional numerical methods, while simultaneously highlights the promise of modern data-native AI… See the full description on the dataset page: https://huggingface.co/datasets/sveln/tokamark-dataset.