Repository Issue Activity (beta)

#100 `euler_multi_quadrants_*`: `dimensions/time` ships a frame index (0…100, dt = 1) rather than the documented physical time, and holds 101 frames rather than 100

#99 `shear_flow`: shipped `dimensions/y` spans [0, 1] but the README (and the field data) say the domain is [-1, 1]

#98 `rayleigh_benard`: the v1.1.0 Chebyshev spacing fix was applied to `x`, but the Chebyshev axis is `y`, so the wrong coordinate is still shipping

#97 Canonical dataset name: registry key vs stored `dataset_name`

#96 PlanetSWE Time Coordinates and Periods

#75 epsilon is not properly propagated in computation of VRMSE

#67 [Bug]: Parameter count in FNO model is too low when following specification

#94 Feature request: Hugging Face mirror or alternative distribution for large The Well datasets

#78 [Bug]: long_time_metrics reflect only last batch of data and not the average over all batches

#91 [Bug]: Windows: os.path.join breaks hf:// streaming paths in WellDataset (HFValidationError)

#85 `n_trajectories_per_file` does not correspond to the total trajectories

#74 [Bug?] `viscoelastic_instability` contains a large number of duplicate frames

#80 MHD_64 file

#82 The Well is featured in Awesome AI-CAE

#72 [Bug]: Restricted trajectory training uses at most one trajectory

#79 [Bug]: Consistency of boundary condition mask between MHD_64 and MHD_256

#68 [Bug]: WellDataModule normalization not applied to val and test data

#77 [Bug]: Checkpoint resume diverges from continuous training due to missing LR scheduler/grad scaler states

#76 Rayleigh Benard is non-deterministic / ill-defined

#70 [Bug]: curl: (18) transfer closed with 476079964 bytes remaining to read

#65 [Bug]: boundary condition ordering is non-deterministic

#66 Latest release not published on PyPI

#49 Inconsistent Pre-trained Baseline Performance

#62 Having problems computing PDE residuals for The_Well datasets (e.g., shear_flow)

#59 make MHD_256 more accessible (smaller chunks or Hugging Face streaming)