Update README.md

README.md

@@ -456,3 +456,337 @@ pretty_name: ConStellaration
 size_categories:
 - 10K<n<100K
 ---
+# Dataset Card for ConStellaration
+
+<!-- Provide a quick summary of the dataset. -->
+
+A dataset of diverse quasi-isodynamic (QI) stellarator boundary shapes with corresponding performance metrics and ideal magneto-hydrodynamic (MHD) equilibria, as well as settings for their generation.
+
+## Dataset Details
+
+### Dataset Description
+
+<!-- Provide a longer summary of what this dataset is. -->
+
+Stellarators are magnetic confinement devices that are being pursued to deliver steady-state carbon-free fusion energy. Their design involves a high-dimensional, constrained optimization problem that requires expensive physics simulations and significant domain expertise. Specifically, QI-stellarators are seen as a promising path to commercial fusion due to their intrinsic avoidance of current-driven disruptions. 
+With the release of this dataset, we aim to lower the barrier for optimization and machine learning researchers to contribute to stellarator design, and to accelerate cross-disciplinary progress toward bringing fusion energy to the grid.
+
+- **Curated by:** Proxima Fusion
+- **License:** MIT
+
+### Dataset Sources
+
+<!-- Provide the basic links for the dataset. -->
+
+- **Repository:** https://huggingface.co/datasets/proxima-fusion/constellaration
+- **Paper:** [not published yet]
+- **Code:** https://github.com/proximafusion/constellaration
+
+## Uses
+
+## Basic Usage
+Load the dataset and convert to a Pandas Dataframe:
+```python
+import datasets
+
+full_flat_ds = datasets.load_dataset("proxima-fusion/constellaration", "full_flat")["train"]
+full_flat_df = full_flat_ds.to_pandas().set_index("plasma_config_id")
+
+for column in full_flat_df.columns:
+    print(column)
+```
+<details>
+<summary>Output</summary>
+
+```python
+boundary.n_field_periods
+boundary.is_stellarator_symmetric
+boundary.r_cos(0, 0)
+boundary.r_cos(0, 1)
+boundary.r_cos(0, 2)
+boundary.r_cos(0, 3)
+boundary.r_cos(0, 4)
+boundary.r_cos(1, -4)
+boundary.r_cos(1, -3)
+boundary.r_cos(1, -2)
+boundary.r_cos(1, -1)
+boundary.r_cos(1, 0)
+boundary.r_cos(1, 1)
+boundary.r_cos(1, 2)
+boundary.r_cos(1, 3)
+boundary.r_cos(1, 4)
+boundary.r_cos(2, -4)
+boundary.r_cos(2, -3)
+boundary.r_cos(2, -2)
+boundary.r_cos(2, -1)
+boundary.r_cos(2, 0)
+boundary.r_cos(2, 1)
+boundary.r_cos(2, 2)
+boundary.r_cos(2, 3)
+boundary.r_cos(2, 4)
+boundary.r_cos(3, -4)
+boundary.r_cos(3, -3)
+boundary.r_cos(3, -2)
+boundary.r_cos(3, -1)
+boundary.r_cos(3, 0)
+boundary.r_cos(3, 1)
+boundary.r_cos(3, 2)
+boundary.r_cos(3, 3)
+boundary.r_cos(3, 4)
+boundary.r_cos(4, -4)
+boundary.r_cos(4, -3)
+boundary.r_cos(4, -2)
+boundary.r_cos(4, -1)
+boundary.r_cos(4, 0)
+boundary.r_cos(4, 1)
+boundary.r_cos(4, 2)
+boundary.r_cos(4, 3)
+boundary.r_cos(4, 4)
+boundary.z_sin(0, 1)
+boundary.z_sin(0, 2)
+boundary.z_sin(0, 3)
+boundary.z_sin(0, 4)
+boundary.z_sin(1, -4)
+boundary.z_sin(1, -3)
+boundary.z_sin(1, -2)
+boundary.z_sin(1, -1)
+boundary.z_sin(1, 0)
+boundary.z_sin(1, 1)
+boundary.z_sin(1, 2)
+boundary.z_sin(1, 3)
+boundary.z_sin(1, 4)
+boundary.z_sin(2, -4)
+boundary.z_sin(2, -3)
+boundary.z_sin(2, -2)
+boundary.z_sin(2, -1)
+boundary.z_sin(2, 0)
+boundary.z_sin(2, 1)
+boundary.z_sin(2, 2)
+boundary.z_sin(2, 3)
+boundary.z_sin(2, 4)
+boundary.z_sin(3, -4)
+boundary.z_sin(3, -3)
+boundary.z_sin(3, -2)
+boundary.z_sin(3, -1)
+boundary.z_sin(3, 0)
+boundary.z_sin(3, 1)
+boundary.z_sin(3, 2)
+boundary.z_sin(3, 3)
+boundary.z_sin(3, 4)
+boundary.z_sin(4, -4)
+boundary.z_sin(4, -3)
+boundary.z_sin(4, -2)
+boundary.z_sin(4, -1)
+boundary.z_sin(4, 0)
+boundary.z_sin(4, 1)
+boundary.z_sin(4, 2)
+boundary.z_sin(4, 3)
+boundary.z_sin(4, 4)
+metrics.qi
+metrics.vacuum_well
+metrics.aspect_ratio
+metrics.max_elongation
+metrics.average_triangularity
+metrics.axis_magnetic_mirror_ratio
+metrics.edge_magnetic_mirror_ratio
+metrics.aspect_ratio_over_edge_rotational_transform
+metrics.flux_compression_in_regions_of_bad_curvature
+metrics.axis_rotational_transform_over_n_field_periods
+metrics.edge_rotational_transform_over_n_field_periods
+metrics.minimum_normalized_magnetic_gradient_scale_length
+metrics.id
+omnigenous_field_and_targets.aspect_ratio
+omnigenous_field_and_targets.major_radius
+omnigenous_field_and_targets.max_elongation
+omnigenous_field_and_targets.rotational_transform
+omnigenous_field_and_targets.id
+omnigenous_field_and_targets.omnigenous_field.x_lmn
+omnigenous_field_and_targets.omnigenous_field.n_field_periods
+omnigenous_field_and_targets.omnigenous_field.poloidal_winding
+omnigenous_field_and_targets.omnigenous_field.torodial_winding
+omnigenous_field_and_targets.omnigenous_field.modB_spline_knot_coefficients
+desc_omnigenous_field_optimization_settings.id
+desc_omnigenous_field_optimization_settings.objective_settings.elongation_settings.name
+desc_omnigenous_field_optimization_settings.objective_settings.elongation_settings.weight
+desc_omnigenous_field_optimization_settings.objective_settings.elongation_settings.target_kind
+desc_omnigenous_field_optimization_settings.objective_settings.omnigenity_settings.name
+desc_omnigenous_field_optimization_settings.objective_settings.omnigenity_settings.weight
+desc_omnigenous_field_optimization_settings.objective_settings.omnigenity_settings.eta_weight
+desc_omnigenous_field_optimization_settings.objective_settings.omnigenity_settings.eq_lcfs_grid_rho
+desc_omnigenous_field_optimization_settings.objective_settings.omnigenity_settings.eq_lcfs_grid_M_factor
+desc_omnigenous_field_optimization_settings.objective_settings.omnigenity_settings.eq_lcfs_grid_N_factor
+desc_omnigenous_field_optimization_settings.objective_settings.aspect_ratio_settings.name
+desc_omnigenous_field_optimization_settings.objective_settings.aspect_ratio_settings.weight
+desc_omnigenous_field_optimization_settings.objective_settings.aspect_ratio_settings.target_kind
+desc_omnigenous_field_optimization_settings.objective_settings.rotational_transform_settings.name
+desc_omnigenous_field_optimization_settings.objective_settings.rotational_transform_settings.weight
+desc_omnigenous_field_optimization_settings.objective_settings.rotational_transform_settings.target_kind
+desc_omnigenous_field_optimization_settings.optimizer_settings.name
+desc_omnigenous_field_optimization_settings.optimizer_settings.maxiter
+desc_omnigenous_field_optimization_settings.optimizer_settings.verbose
+desc_omnigenous_field_optimization_settings.equilibrium_settings.psi
+desc_omnigenous_field_optimization_settings.equilibrium_settings.check_orientation
+desc_omnigenous_field_optimization_settings.equilibrium_settings.max_poloidal_mode
+desc_omnigenous_field_optimization_settings.equilibrium_settings.max_toroidal_mode
+desc_omnigenous_field_optimization_settings.initial_guess_settings.torsion
+desc_omnigenous_field_optimization_settings.initial_guess_settings.elongation
+desc_omnigenous_field_optimization_settings.initial_guess_settings.aspect_ratio
+desc_omnigenous_field_optimization_settings.initial_guess_settings.major_radius
+desc_omnigenous_field_optimization_settings.initial_guess_settings.mirror_ratio
+desc_omnigenous_field_optimization_settings.initial_guess_settings.n_field_periods
+desc_omnigenous_field_optimization_settings.initial_guess_settings.is_iota_positive
+desc_omnigenous_field_optimization_settings.initial_guess_settings.is_stellarator_symmetric
+desc_omnigenous_field_optimization_settings.initial_guess_settings.max_elongation
+desc_omnigenous_field_optimization_settings.initial_guess_settings.max_poloidal_mode
+desc_omnigenous_field_optimization_settings.initial_guess_settings.max_toroidal_mode
+desc_omnigenous_field_optimization_settings.initial_guess_settings.rotational_transform
+vmec_omnigenous_field_optimization_settings.verbose
+vmec_omnigenous_field_optimization_settings.max_poloidal_mode
+vmec_omnigenous_field_optimization_settings.max_toroidal_mode
+vmec_omnigenous_field_optimization_settings.n_inner_optimizations
+vmec_omnigenous_field_optimization_settings.gradient_free_max_time
+vmec_omnigenous_field_optimization_settings.infinity_norm_spectrum_scaling
+vmec_omnigenous_field_optimization_settings.gradient_free_budget_per_design_variable
+vmec_omnigenous_field_optimization_settings.use_continuation_method_in_fourier_space
+vmec_omnigenous_field_optimization_settings.gradient_free_optimization_hypercube_bounds
+vmec_omnigenous_field_optimization_settings.gradient_based_relative_objectives_tolerance
+vmec_omnigenous_field_optimization_settings.id
+qp_init_omnigenous_field_optimization_settings.torsion
+qp_init_omnigenous_field_optimization_settings.elongation
+qp_init_omnigenous_field_optimization_settings.aspect_ratio
+qp_init_omnigenous_field_optimization_settings.major_radius
+qp_init_omnigenous_field_optimization_settings.mirror_ratio
+qp_init_omnigenous_field_optimization_settings.n_field_periods
+qp_init_omnigenous_field_optimization_settings.is_iota_positive
+qp_init_omnigenous_field_optimization_settings.is_stellarator_symmetric
+qp_init_omnigenous_field_optimization_settings.id
+nae_init_omnigenous_field_optimization_settings.aspect_ratio
+nae_init_omnigenous_field_optimization_settings.mirror_ratio
+nae_init_omnigenous_field_optimization_settings.max_elongation
+nae_init_omnigenous_field_optimization_settings.n_field_periods
+nae_init_omnigenous_field_optimization_settings.max_poloidal_mode
+nae_init_omnigenous_field_optimization_settings.max_toroidal_mode
+nae_init_omnigenous_field_optimization_settings.rotational_transform
+nae_init_omnigenous_field_optimization_settings.id
+misc.vmecpp_wout_id
+misc.has_optimize_boundary_omnigenity_vmec_error
+misc.has_optimize_boundary_omnigenity_desc_error
+misc.has_generate_qp_initialization_from_targets_error
+misc.has_generate_nae_initialization_from_targets_error
+misc.has_neurips_2025_forward_model_error
+```
+</details>
+
+## Advanced Usage
+Install `constellaration` from [here](https://github.com/proximafusion/constellaration).
+
+Load and instantiate plasma boundaries:
+```python
+import datasets
+from constellaration.geometry import surface_rz_fourier
+
+full_json_ds = datasets.load_dataset("proxima-fusion/constellaration", "full_json", streaming=True)["train"]
+
+for batch in full_json_ds.iter(batch_size=1):
+    boundary_json = batch["boundary.surface"][0]
+    boundary = surface_rz_fourier.SurfaceRZFourier.model_validate_json(boundary_json)
+    break
+
+print(boundary)
+```
+Stream and instantiate the VMEC ideal MHD equilibria:
+```python
+import datasets
+from research.neurips_2025.open_source.mhd import vmec_utils
+
+ds = datasets.load_dataset("proxima-fusion/constellaration", "vmecpp_ideal_mhd_equilibria", streaming=True)["train"]
+
+for batch in ds.filter(lambda row: row["vmecpp_wout_json"] is not None).iter(batch_size=1):
+    vmecpp_wout_json = batch["vmecpp_wout_json"][0]
+    vmecpp_wout = vmec_utils.VmecppWOut.model_validate_json(vmecpp_wout_json)
+    break
+
+print(vmecpp_wout.n_field_periods)
+```
+
+## Dataset Structure
+
+<!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->
+
+The dataset consists of 2 tabular parts. Both parts have a column `plasma_config_id` in common which can be used to associate respective entries.
+
+### 1. `full_json`/`full_flat`
+Contains information about:
+- plasma boundaries  
+- ideal MHD metrics
+- omnigenous field and targets, used as input for sampling of plasma boundaries
+- sampling settings for various methods ([DESC](https://desc-docs.readthedocs.io/en/stable/), [VMEC](https://github.com/proximafusion/vmecpp), QP initialization, Near-axis expansion)
+- Miscellaneous information about errors that might have occurred during sampling or metrics computation.
+
+The `full_json` variant of the dataset contains for each of the components listed about an identifier column (ending with `.id`), as well as a JSON column.
+
+The `full_flat` variant contains the same information as `full_json` but with all JSON columns flattened into one column per leaf in the nested JSON structure (with `.` separating the keys on the JSON path to the respective leaf).
+
+### 2. `vmecpp_ideal_mhd_equilibria`
+Contain for each plasma boundary a JSON representations of the "WOut" file as obtained when running VMEC++ initialized on the boundary.
+The JSON representation can be converted to a VMEC2000 output file.
+
+## Dataset Creation
+
+### Curation Rationale
+
+<!-- Motivation for the creation of this dataset. -->
+
+Wide-spread community progress is currently bottlenecked by the lack of standardized optimization problems with strong baselines and datasets that enable data-driven approaches, particularly for quasi-isodynamic (QI) stellarator configurations.
+
+### Source Data
+
+<!-- This section describes the source data (e.g. news text and headlines, social media posts, translated sentences, ...). -->
+
+#### Data Collection and Processing
+
+<!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->
+
+We generated this dataset by sampling diverse QI fields and optimizing stellarator plasma boundaries to target key properties, using four different methods.
+
+#### Who are the source data producers?
+
+<!-- This section describes the people or systems who originally created the data. It should also include self-reported demographic or identity information for the source data creators if this information is available. -->
+
+Proxima Fusion's stellarator optimization team.
+
+#### Personal and Sensitive Information
+
+<!-- State whether the dataset contains data that might be considered personal, sensitive, or private (e.g., data that reveals addresses, uniquely identifiable names or aliases, racial or ethnic origins, sexual orientations, religious beliefs, political opinions, financial or health data, etc.). If efforts were made to anonymize the data, describe the anonymization process. -->
+
+The dataset contains no personally identifiable information.
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the dataset, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[not published yet]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the dataset or dataset card. -->
+
+| Abbreviation | Expansion |
+| -------- | ------- |
+| QI | Quasi-Isodynamic(ity) |
+| MHD | Magneto-Hydrodynamic |
+| DESC | Dynamical Equilibrium Solver for Confinement |
+| VMEC | Variational Moments Equilibrium Code |
+| QP | Quasi-Poloidal |
+| NAE | Near-Axis Expansion |
+| NFP | Number of Field Periods |
+
+## Dataset Card Authors [optional]
+
+Alexander Bauer, Santiago A. Cadena
+
+## Dataset Card Contact
+
+[email protected]