Adding derivative training - Githubissues

hippylib / hippyflow

Dimension reduced surrogate construction for parametric PDE maps

Other

36 stars 6 forks source link

Adding derivative training #4

Closed tomoleary closed 3 years ago

tomoleary commented 3 years ago

Generate low rank Jacobian training data
Add h1 low-rank semi-norm loss to objective (in hessianlearn)
Derivative based training