Dataset Viewer
Duplicate
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
index: int64
name: string
energy_source: string
energy_reference: string
J_natoms: int64
J_energy_eV: double
J_energy_per_atom_eV: double
J_species: list<item: string>
  child 0, item: string
electronic_converged: bool
OUTCAR_sha256: string
INCAR: struct<ALGO: string, EDIFF: double, EDIFFG: double, ENCUT: double, IBRION: int64, ISIF: int64, ISMEA (... 219 chars omitted)
  child 0, ALGO: string
  child 1, EDIFF: double
  child 2, EDIFFG: double
  child 3, ENCUT: double
  child 4, IBRION: int64
  child 5, ISIF: int64
  child 6, ISMEAR: int64
  child 7, ISPIN: int64
  child 8, ISTART: int64
  child 9, LASPH: bool
  child 10, LCHARG: bool
  child 11, LMAXMIX: int64
  child 12, LORBIT: int64
  child 13, LREAL: string
  child 14, LWAVE: bool
  child 15, MAGMOM: list<item: double>
      child 0, item: double
  child 16, NELM: int64
  child 17, NPAR: int64
  child 18, NSW: int64
  child 19, PREC: string
  child 20, SIGMA: double
KPOINTS: struct<@module: string, @class: string, comment: string, nkpoints: int64, generation_style: string,  (... 182 chars omitted)
  child 0, @module: string
  child 1, @class: string
  child 2, comment: string
  child 3, nkpoints: int64
  child 4, generation_style: string
  child 5, kpoints: list<item: list<item: int64>>
      child 0, item: list<item: int64>
          child 0, item: int64
  child 6, usershift: list<item: int64>
      child 0, item: int64
  child 7, kpts_weights: null
  child 8, coord_type: null
  child 9, labels: null
  child 10, tet_num
...
ist<item (... 323 chars omitted)
  child 0, @module: string
  child 1, @class: string
  child 2, charge: double
  child 3, lattice: struct<matrix: list<item: list<item: double>>, pbc: list<item: bool>, a: double, b: double, c: doubl (... 62 chars omitted)
      child 0, matrix: list<item: list<item: double>>
          child 0, item: list<item: double>
              child 0, item: double
      child 1, pbc: list<item: bool>
          child 0, item: bool
      child 2, a: double
      child 3, b: double
      child 4, c: double
      child 5, alpha: double
      child 6, beta: double
      child 7, gamma: double
      child 8, volume: double
  child 4, properties: struct<>
  child 5, sites: list<item: struct<species: list<item: struct<element: string, occu: int64>>, abc: list<item: double> (... 64 chars omitted)
      child 0, item: struct<species: list<item: struct<element: string, occu: int64>>, abc: list<item: double>, propertie (... 52 chars omitted)
          child 0, species: list<item: struct<element: string, occu: int64>>
              child 0, item: struct<element: string, occu: int64>
                  child 0, element: string
                  child 1, occu: int64
          child 1, abc: list<item: double>
              child 0, item: double
          child 2, properties: struct<>
          child 3, label: string
          child 4, xyz: list<item: double>
              child 0, item: double
source_state_structure_id: int64
parent_structure_id: int64
energy_eV: double
to
{'global_index': Value('int64'), 'material': Value('string'), 'formula': Value('string'), 'structure_label': Value('string'), 'spin_label': Value('string'), 'source_frame': Value('int64'), 'input_ionic_step': Value('int64'), 'natoms': Value('int64'), 'mp_id': Value('string'), 'parent_state_id': Value('int64'), 'parent_structure_id': Value('int64'), 'source_state_structure_id': Value('int64'), 'sample_role': Value('string'), 'parent_magnetic_order': Value('string'), 'vasp_exit': Value('int64'), 'success': Value('bool'), 'converged_electronic': Value('bool'), 'converged_ionic': Value('bool'), 'pymatgen_converged_ionic': Value('bool'), 'normal_termination': Value('bool'), 'ionic_steps': Value('int64'), 'last_electronic_steps': Value('int64'), 'reached_nsw_limit': Value('bool'), 'energy_eV': Value('float64'), 'final_structure': {'@module': Value('string'), '@class': Value('string'), 'charge': Value('float64'), 'lattice': {'matrix': List(List(Value('float64'))), 'pbc': List(Value('bool')), 'a': Value('float64'), 'b': Value('float64'), 'c': Value('float64'), 'alpha': Value('float64'), 'beta': Value('float64'), 'gamma': Value('float64'), 'volume': Value('float64')}, 'properties': {}, 'sites': List({'species': List({'element': Value('string'), 'occu': Value('int64')}), 'abc': List(Value('float64')), 'properties': {}, 'label': Value('string'), 'xyz': List(Value('float64'))})}, 'final_magmom_poscar_order': List(Value('float64')), 'experimental_order_retained': Value('bool'), 'energy_source': Value('string'), 'energy_reference': Value('string'), 'label_sources': {'structure_and_magmom': Value('string'), 'energy': Value('string')}, 'exported_record_is_single_calculation': Value('bool')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 149, in get_rows_or_raise
                  return get_rows(
                      dataset=dataset,
                  ...<4 lines>...
                      column_names=column_names,
                  )
                File "/src/libs/libcommon/src/libcommon/utils.py", line 274, in decorator
                  return func(*args, **kwargs)
                File "/src/services/worker/src/worker/utils.py", line 129, in get_rows
                  rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
                File "/src/services/worker/src/worker/utils.py", line 544, in safe_iter
                  yield from ds.decode(False) if ds.features else ds
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 3074, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2578, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2603, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 577, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 459, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 373, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2394, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2322, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              index: int64
              name: string
              energy_source: string
              energy_reference: string
              J_natoms: int64
              J_energy_eV: double
              J_energy_per_atom_eV: double
              J_species: list<item: string>
                child 0, item: string
              electronic_converged: bool
              OUTCAR_sha256: string
              INCAR: struct<ALGO: string, EDIFF: double, EDIFFG: double, ENCUT: double, IBRION: int64, ISIF: int64, ISMEA (... 219 chars omitted)
                child 0, ALGO: string
                child 1, EDIFF: double
                child 2, EDIFFG: double
                child 3, ENCUT: double
                child 4, IBRION: int64
                child 5, ISIF: int64
                child 6, ISMEAR: int64
                child 7, ISPIN: int64
                child 8, ISTART: int64
                child 9, LASPH: bool
                child 10, LCHARG: bool
                child 11, LMAXMIX: int64
                child 12, LORBIT: int64
                child 13, LREAL: string
                child 14, LWAVE: bool
                child 15, MAGMOM: list<item: double>
                    child 0, item: double
                child 16, NELM: int64
                child 17, NPAR: int64
                child 18, NSW: int64
                child 19, PREC: string
                child 20, SIGMA: double
              KPOINTS: struct<@module: string, @class: string, comment: string, nkpoints: int64, generation_style: string,  (... 182 chars omitted)
                child 0, @module: string
                child 1, @class: string
                child 2, comment: string
                child 3, nkpoints: int64
                child 4, generation_style: string
                child 5, kpoints: list<item: list<item: int64>>
                    child 0, item: list<item: int64>
                        child 0, item: int64
                child 6, usershift: list<item: int64>
                    child 0, item: int64
                child 7, kpts_weights: null
                child 8, coord_type: null
                child 9, labels: null
                child 10, tet_num
              ...
              ist<item (... 323 chars omitted)
                child 0, @module: string
                child 1, @class: string
                child 2, charge: double
                child 3, lattice: struct<matrix: list<item: list<item: double>>, pbc: list<item: bool>, a: double, b: double, c: doubl (... 62 chars omitted)
                    child 0, matrix: list<item: list<item: double>>
                        child 0, item: list<item: double>
                            child 0, item: double
                    child 1, pbc: list<item: bool>
                        child 0, item: bool
                    child 2, a: double
                    child 3, b: double
                    child 4, c: double
                    child 5, alpha: double
                    child 6, beta: double
                    child 7, gamma: double
                    child 8, volume: double
                child 4, properties: struct<>
                child 5, sites: list<item: struct<species: list<item: struct<element: string, occu: int64>>, abc: list<item: double> (... 64 chars omitted)
                    child 0, item: struct<species: list<item: struct<element: string, occu: int64>>, abc: list<item: double>, propertie (... 52 chars omitted)
                        child 0, species: list<item: struct<element: string, occu: int64>>
                            child 0, item: struct<element: string, occu: int64>
                                child 0, element: string
                                child 1, occu: int64
                        child 1, abc: list<item: double>
                            child 0, item: double
                        child 2, properties: struct<>
                        child 3, label: string
                        child 4, xyz: list<item: double>
                            child 0, item: double
              source_state_structure_id: int64
              parent_structure_id: int64
              energy_eV: double
              to
              {'global_index': Value('int64'), 'material': Value('string'), 'formula': Value('string'), 'structure_label': Value('string'), 'spin_label': Value('string'), 'source_frame': Value('int64'), 'input_ionic_step': Value('int64'), 'natoms': Value('int64'), 'mp_id': Value('string'), 'parent_state_id': Value('int64'), 'parent_structure_id': Value('int64'), 'source_state_structure_id': Value('int64'), 'sample_role': Value('string'), 'parent_magnetic_order': Value('string'), 'vasp_exit': Value('int64'), 'success': Value('bool'), 'converged_electronic': Value('bool'), 'converged_ionic': Value('bool'), 'pymatgen_converged_ionic': Value('bool'), 'normal_termination': Value('bool'), 'ionic_steps': Value('int64'), 'last_electronic_steps': Value('int64'), 'reached_nsw_limit': Value('bool'), 'energy_eV': Value('float64'), 'final_structure': {'@module': Value('string'), '@class': Value('string'), 'charge': Value('float64'), 'lattice': {'matrix': List(List(Value('float64'))), 'pbc': List(Value('bool')), 'a': Value('float64'), 'b': Value('float64'), 'c': Value('float64'), 'alpha': Value('float64'), 'beta': Value('float64'), 'gamma': Value('float64'), 'volume': Value('float64')}, 'properties': {}, 'sites': List({'species': List({'element': Value('string'), 'occu': Value('int64')}), 'abc': List(Value('float64')), 'properties': {}, 'label': Value('string'), 'xyz': List(Value('float64'))})}, 'final_magmom_poscar_order': List(Value('float64')), 'experimental_order_retained': Value('bool'), 'energy_source': Value('string'), 'energy_reference': Value('string'), 'label_sources': {'structure_and_magmom': Value('string'), 'energy': Value('string')}, 'exported_record_is_single_calculation': Value('bool')}
              because column names don't match

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

MagBench data

Benchmark data of MagBench, a five-dimensional benchmark for magnetic interatomic potentials: magnetic ground state, energy above the Materials Project hull, phonon spectra, exchange parameters J, and the volume dependence of J, on 191 magnetic materials.

The evaluation code (runner and scorer) is in the code repository: https://github.com/kairosmaterial/Magbench.

Data version: v0.2.2. Scoring protocol: v0.2.1. See CHANGELOG.

Layout

Path Materials Content
inputs/ — Everything a model receives: structures, sign-pattern references, exchange bonds, energy-mapping designs, phonon sampling
targets/ — DFT reference values and the scoring configuration (scoring.json)
results/prophet_spin_v1/ 191 Reference scores of Prophet-Spin v1 (total 0.43731 under protocol v0.2.1)
01_refined_structures_191/ 191 Final structures of the original DFT relaxations (EXTXYZ), experimental and DFT moments, MP hull labels, provenance
02_magnetovolume_derivatives_133/ 133 Volume-strain calculations and reference slopes ∂J/∂ln V
03_phonons_110/ 110 Displaced supercells, reference sampling points, DFT frequencies and force constants
04_exchange_J_172/ 172 Exchange bonds, shells, energy-mapping designs and DFT J
DFT_reference.md — KPOINTS and POTCAR settings of the reference calculations
hull_potcar_metadata.json 191 Per-material POTCAR titles, dates and checksums

inputs/ is used for model inference and targets/ for scoring. targets/structures.extxyz contains the 191 final structures of the original DFT relaxations with both sets of moments. The reference energy labels for the MP hull are in targets/hull.json; the energies are taken from the static reference states of the J calculations and use the fixed MP 2023-02-07 GGA/GGA+U hull.

Usage

huggingface-cli download kairosmaterial/Magbench --repo-type dataset --local-dir data
python -m pip install ./benchmark_runner          # from the code repository
magbench score --targets data/targets --predictions /path/to/predictions.json --output scores.json

The scoring rules of the five dimensions are described in the README of the code repository.

Downloads last month
-