entries_model#
- t3toolbox.fitting.entries_model(geometry, x, index, residual, regularizer=None)#
def entries_model( geometry, # MANIFOLD / COREWISE (or the UNIFORM_* twin for a uniform x) x: typ.Union[t3.TuckerTensorTrain, ut3.UniformTuckerTensorTrain], # the current point index: NDArray, # int, shape=(d,)+W -- the grid points residual: NDArray, # r = entries(x) − y, shape W+C regularizer: typ.Any = None, # optional regularizer, e.g. optimizers.IdentityRegularizer(λ) ) -> typ.Union[GaussNewtonModel, UniformGaussNewtonModel]:
The Gauss-Newton model of an all-modes
entriesleast-squares objective atx, ongeometry.Identical to
apply_model()but the measurements are tensor entries at integer grid pointsindex(shape(d,)+W) rather than applies against probe vectors.- Parameters:
x (t3toolbox.backend.common.typ.Union[TuckerTensorTrain, UniformTuckerTensorTrain])
index (NDArray)
residual (NDArray)
regularizer (t3toolbox.backend.common.typ.Any)
- Return type:
t3toolbox.backend.common.typ.Union[GaussNewtonModel, UniformGaussNewtonModel]