# The Well: Viscoelastic Instability and Elasto-Inertial Turbulence

## Facts

- Provider: Polymathic AI (The Well); Miguel Beneitez and Richard Kerswell, DAMTP, University of Cambridge
- License: CC-BY-4.0 (https://arxiv.org/html/2412.00568)
- Size: 29 files
- Format: HDF5 (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/polymathic-ai/viscoelastic_instability
- Access: public, readable without an account
- Category: Physics
- Source: https://arxiv.org/abs/2412.00568
- Page: https://scigantic.com/archive/40e8cdf0-1beb-5d3b-8aea-f6179839bc88
- Keywords: Physics, HDF5, Hugging Face

## About

2D direct numerical simulations of channel flow of a dilute polymer solution, showing multistability: up to four coexisting attractors for the same parameters, namely laminar flow, a steady arrowhead regime (SAR), elasto-inertial turbulence (EIT) and a chaotic arrowhead regime (CAR). Solved with Dedalus using the FENE-P model at Reynolds number 1000, Weissenberg number 50, beta 0.9 and L_max 70 on 512 by 512 grids, with no-slip walls. Fields are pressure, velocity and the polymer conformation tensor. The datasheet lists 260 trajectories, including 34 EIT, 39 CAR and 20 SAR runs and snapshots of transitions to chaos. Listing data/ (measured 2026-11-13) shows train/ (7 files, 56.4 GB, one per regime such as AH, CAR and EIT), valid/ and test/ (7 files, 5.9 GB each); about 68 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.

## Load the data

```python
from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="polymathic-ai/viscoelastic_instability", filename="data/test/viscoelastic_instability_AH.hdf5", repo_type="dataset")
print(path)
```

## Data structure

File types in sampled files: HDF5

Example file: `data/test/viscoelastic_instability_AH.hdf5`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/40e8cdf0-1beb-5d3b-8aea-f6179839bc88
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/40e8cdf0-1beb-5d3b-8aea-f6179839bc88
- Over MCP: `get_schema_card` with id `40e8cdf0-1beb-5d3b-8aea-f6179839bc88` at https://api.scigantic.com/mcp

## Related topics

- [HDF5 datasets](https://scigantic.com/topics/hdf5)
- [Physics datasets](https://scigantic.com/topics/physics)
