{"success":true,"data":{"author":"Polymathic AI (The Well); Jonah Miller, Los Alamos National Laboratory","bucketAccessType":"public","bucketName":"polymathic-ai/post_neutron_star_merger","bucketPrefix":"","category":"Physics","cloudProvider":"huggingface","createdAt":"2026-11-14T01:01:19.620Z","description":"Simulations of the aftermath of a neutron star merger: an accretion torus around a black hole, from full 3D general relativistic neutrino radiation magnetohydrodynamics computed with the open-source code nubhlight (LANL), stored as axisymmetrized snapshots. Plasma is evolved with finite volumes and constrained transport on a curvilinear log-spherical grid, closed by a tabulated nuclear equation of state, with neutrinos handled by Monte Carlo transport (visible here only through source terms). Each trajectory has 181 steps of 192 by 128 by 66 cells. Fields are density, internal energy, electron fraction, temperature, entropy, velocity, magnetic field and the spacetime metric. The datasheet lists 8 simulations varying black hole spin and mass and the torus mass, angular momentum, electron fraction and entropy. Listing data/ (measured 2026-11-13) shows train/ (6 files), valid/ (1 file) and test/ (1 file), one 14.1 GB scenario per file; about 113 GB in the repository. Files are HDF5 in the common The Well format; open them with the_well package (WellDataModule with base path hf://datasets/polymathic-ai/) or directly with h5py. Part of The Well (Ohana et al., NeurIPS 2024 Datasets and Benchmarks, arXiv:2412.00568); released under CC-BY-4.0, cite The Well and the simulation's source papers.","displayTitle":"The Well: Post Neutron Star Merger Accretion Disk (GR Neutrino Radiation MHD)","id":"ba3d411d-083b-5bf8-81bb-0da1d38420ee","license":"CC-BY-4.0","licenseEvidenceUrl":"https://arxiv.org/html/2412.00568","objectCountEstimate":17,"probeStatus":"ok","referenceUrl":"https://arxiv.org/abs/2412.00568","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"polymathic-ai/post_neutron_star_merger\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"polymathic-ai/post_neutron_star_merger\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=polymathic-ai, license:cc-by-4.0, downloads=457"],"format":"mixed","generatedAt":"2026-11-14T01:01:24.098Z","layout":{"example":"data/test/post_neutron_star_merger_scenario_2.hdf5","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"data/test/post_neutron_star_merger_scenario_2.hdf5","truncatedAt":512}],"sampleObjects":["data/test/post_neutron_star_merger_scenario_2.hdf5","data/train/post_neutron_star_merger_scenario_0.hdf5","data/train/post_neutron_star_merger_scenario_3.hdf5","data/train/post_neutron_star_merger_scenario_4.hdf5","data/train/post_neutron_star_merger_scenario_5.hdf5"],"sidecars":[{"bytes":"---\nlanguage:\n- en\nlicense: cc-by-4.0\ntags:\n- physics\n\ntask_categories:\n- time-series-forecasting\n- other\ntask_ids:\n- multivariate-time-series-forecasting\n---\n\nThis Dataset is part of <a href=\"arxiv.org/abs/2412.00568\">The Well Collection</a>.\n\n# How To Load from HuggingFace Hub\n\n1. Be sure to have `the_well` installed (`pip install the_well`)\n2. Use the `WellDataModule` to retrieve data as follows:\n\n```python\nfrom the_well.benchmark.data import WellDataModule\n\n# The following line may take a couple of minutes to instantiate the datamodule\ndatamodule = WellDataModule(\n    \"hf://datasets/polymathic-ai/\",\n    \"post_neutron_star_merger\",\n)\ntrain_dataloader = datamodule.train_dataloader()\n\nfor batch in dataloader:\n    # Process training batch\n    ...\n```\n\n# Post neutron star merger\n\n**One line description of the data:** Simulations of the aftermath of a neutron star merger.\n\n**Longer description of the data:** The simulations presented here are axisymmetrized snapshots of full three-dimensional general relativistic neutrino radiation magnetohydrodynamics. The plasma physics is treated with finite volumes with constrained transport for the magnetic field on a curvilinear grid. The system is closed by a tabulated nuclear equation of state assuming nuclear statistical equilibrium (NSE). The radiation field is treated via Monte Carlo transport, which is a particle method. The particles are not included in this dataset, however their effects are visible as source terms on the fluid.\n\n**Associated paper**: The simulations included here are from a series of papers: [Paper 1](https://iopscience.iop.org/article/10.3847/1538-4365/ab09fc/pdf), [Paper 2](https://link.aps.org/accepted/10.1103/PhysRevD.100.023008), [Paper 3](https://arxiv.org/abs/1912.03378), [Paper 4](https://arxiv.org/abs/2212.10691), [Paper 5](https://arxiv.org/abs/2311.05796).\n\n**Domain expert**: [Jonah Miller](https://www.thephysicsmill.com/), Los Alamos National Laboratory.\n\n**Code or software used to generate the data**: Open source software [nublight](https://github.com/lanl/nubhlight).\n\n**Equation**: See equations 1-5 and 16 of Miller, Ryan, Dolence (2019).\n\nThe fluid sector consists of the following system of equations.\n\n$$\n\\begin{align}\n  \\partial_t \\left(\\sqrt{g}\\rho_0 u^t\\right) + \\partial_i\\left(\\sqrt{g}\\rho_0u^i\\right)\n  &= 0\\\\\n  \\partial_t\\left[\\sqrt{g} \\left(T^t_{\\ \\nu} + \\rho_0u^t \\delta^t_\\nu\\right)\\right]\n  + \\partial_i\\left[\\sqrt{g}\\left(T^i_{\\ \\nu} + \\rho_0 u^i \\delta^t_\\nu\\right)\\right]\n  &= \\sqrt{g} \\left(T^\\kappa_{\\ \\lambda} \\Gamma^\\lambda_{\\nu\\kappa} + G_\\nu\\right)\\,\\,\\,\\, \\forall \\nu = 0,1,\\ldots,4\\\\\n  \\partial_t \\left(\\sqrt{g} B^i\\right) + \\partial_j \\left[\\sqrt{g}\\left(b^ju^i - b^i u^j\\right)\\right] &= 0\\\\\n  \\partial_t\\left(\\sqrt{g}\\rho_0 Y_e u^t\\right) + \\partial_i\\left(\\sqrt{g}\\rho_0Y_eu^i\\right)\n  &= \\sqrt{g} G_{\\text{ye}}\\\\\n\\end{align}\n$$\n\nThe standard radiative transfer equation is\n$$\n\\begin{equation}\n    \\frac{D}{d\\lambda}\\left(\\frac{h^3\\mathcal{I}_{\\nu,f}}{\\varepsilon^3}\\right) = \\left(\\frac{h^2\\eta_{\\nu,f}}{\\varepsilon^2}\\right) - \\left(\\frac{\\varepsilon \\chi_{\\nu,f}}{h}\\right) \\left(\\frac{h^3\\mathcal{I}_{\\nu,f}}{\\varepsilon^3}\\right),\n\\end{equation}\n$$\n\n<p align=\"center\"> <img src=\"https://users.flatironinstitute.org/~polymathic/data/the_well/datasets/post_neutron_star_merger/gif/Ye_good_normalized.gif\" width=\"50%\"></p>\n\n\n\n\n| Dataset    | FNO | TFNO  | Unet | CNextU-net\n|:-:|:-:|:-:|:-:|:-:|\n| `post_neutron_star_merger`  | 0.3866 | \\\\(\\mathbf{0.3793}\\\\) | - |-|\n\nTable: VRMSE metrics on test sets (lower is better). Best results are shown in bold. VRMSE is scaled such that predicting the mean value of the target field results in a score of 1. Unet and CNextU-net results are not available as these architectures require all the dimensions of the data to be multiples of 2.\n\n\n# About the data\n\n**Dimension of discretized data:** 181 time-steps of 192 \\\\(\\times\\\\) 128 \\\\(\\times\\\\) 66 snapshots.\n\n**Fields available in the data:** fluid density (scalar field), fluid internal energy (scalar field), electron fraction (scalar field), temperate (sc","encoding":"utf8","key":"README.md","name":"README.md","size":14530,"truncated":true}],"sourceObjectKey":"data/test/post_neutron_star_merger_scenario_2.hdf5","starterCell":"from huggingface_hub import hf_hub_download\npath = hf_hub_download(repo_id=\"polymathic-ai/post_neutron_star_merger\", filename=\"data/test/post_neutron_star_merger_scenario_2.hdf5\", repo_type=\"dataset\")\nprint(path)","starterSamplePath":"data/test/post_neutron_star_merger_scenario_2.hdf5","totalObjects":17},"summary":"Simulations of the aftermath of a neutron star merger: an accretion torus around a black hole, from full 3D general relativistic neutrino radiation magnetohydrodynamics computed…","title":"The Well: Post Neutron Star Merger Accretion Disk (GR Neutrino Radiation MHD)","updatedAt":"2026-11-14T02:36:25.759Z","visibility":"public","websiteUrl":"https://polymathic-ai.org/the_well/"}}