Input–output pairs for the 1D viscous Burgers' equation on a periodic
domain, intended for training and benchmarking neural operators.
Each sample maps an initial condition u_0(x) = u(x, 0) to the solution
u(x, t_end). Initial conditions are drawn from a Gaussian random
field ande the PDE is solved numerically with a pseudo-spectral method.
The 1D viscous Burgers' equation on a periodic domain $x \in [0, 1)$ reads
∂u∂t=−u∂u∂x+ν∂2u∂x2,
\frac{\partial… See the full description on the dataset page:
https://huggingface.co/datasets/TortillaChip/burgers1d-periodic.
