hierarchy¶

pymor-demo hierarchy [OPTIONS] PROBLEM_NUMBER REGRESSOR GRID_INTERVALS NUM_PARAMETERS

Adaptive model hierarchy combining reduced basis and machine learning methods.

Problem number 0 considers an elliptic problem and problem number 1 considers a parabolic problem.

Arguments:

PROBLEM_NUMBER: Selects the problem to solve [0 or 1]. [Required, Choices: 0, 1]
REGRESSOR: Regressor to use. Options are neural networks using PyTorch, pyMOR’s VKOGA algorithm or Gaussian process regression using scikit-learn. [Required, Choices: fcnn, vkoga, gpr]
GRID_INTERVALS: Grid interval count. [Required]
NUM_PARAMETERS: Number of parameters to evaluate the hierarchy for. [Required]

Parameters:

--time-steps: Number of time steps used for discretization (only used if problem_number is 1). [Default: 10]
--time-vectorized, --no-time-vectorized: Predict the whole time trajectory at once or iteratively. [Default: True]
--vis, --no-vis: Visualize estimated errors for the queried parameters. [Default: False]
--validation-ratio: Ratio of training data used for validation of the neural networks. [Default: 0.1]
--input-scaling, --no-input-scaling: Scale the input of the regressor (i.e. the parameter). [Default: False]
--output-scaling, --no-output-scaling: Scale the output of the regressor (i.e. reduced coefficients or output quantity). [Default: False]
--use-dd-model, --no-use-dd-model: Include the data-driven surrogate in the hierarchy. If False, the hierarchy consists only of the reduced basis model and the full-order model. [Default: True]
--quantity: Quantity of interest to query and certify: the 'state' (the solution), the 'output', or 'state+output'. The hierarchy selects the model based on the error estimator(s) matching the requested quantity; for 'state+output' a model is only accepted if both estimates are below the tolerance. [Choices: state, output, state+output, Default: state]
--time-reduction: How the time-dependent error estimate of an instationary problem is reduced to a single number compared against the tolerance: 'max' (maximum in time) or 'mean' (mean in time). [Choices: max, mean, Default: mean]