mud_examples.linear package#
Submodules#
mud_examples.linear.lin module#
- mud_examples.linear.lin.compare_linear_sols(transform, lam_ref, A, b, alpha=1, mean=None, cov=None)[source]#
Input dimension fixed, varying according to the output of the anonymous function transform’s return.
- mud_examples.linear.lin.compare_linear_sols_dim(lam_ref, A, b, alpha=1, mean=None, cov=None)[source]#
Input dimension fixed, varying output dimension.
- mud_examples.linear.lin.compare_linear_sols_rank_list(lam_ref, A, b, alpha=1, mean=None, cov=None)[source]#
Input and output dimensions fixed, varying rank 1..dim_output.
- mud_examples.linear.lin.contour_example(A=array([[1, 1]]), b=array([[0.]]), cov_11=0.5, cov_01=-0.25, initial_mean=array([0.25, 0.25]), alpha=1, omega=1, obs_std=1, show_full=True, show_data=True, show_est=False, param_ref=None, compare=False, fsize=42, figname='latest_figure.png', save=False)[source]#
alpha: float in [0, 1], weight of Tikhonov regularization omega: float in [0, 1], weight of Modified regularization
- mud_examples.linear.lin.main_contours(args)[source]#
Main entrypoint for 2D Linear Rank-Deficient Example (Contour Plots)
- mud_examples.linear.lin.main_dim(args)[source]#
Main entrypoint for High-Dim Linear Dimension Example
- mud_examples.linear.lin.main_meas(args)[source]#
Main entrypoint for High-Dim Linear Measurement Example
- mud_examples.linear.lin.main_meas_var(args)[source]#
Main entrypoint for High-Dim Linear Measurement Example
- mud_examples.linear.lin.setup_logging(loglevel)[source]#
Setup basic logging
- Parameters:
loglevel (int) – minimum loglevel for emitting messages
mud_examples.linear.models module#
- mud_examples.linear.models.createNoisyReferenceData(M, reference_point, std, num_data=None)[source]#
- mud_examples.linear.models.createRandomLinearMap(dim_input, dim_output, dist='normal', repeated=False)[source]#
Create random linear map from P dimensions to S dimensions.
- mud_examples.linear.models.createRandomLinearPair(reference_point, num_data, std, dist='normal', repeated=False)[source]#
data will come from a normal distribution centered at zero with standard deviation given by std QoI map will come from standard uniform, or normal if dist=normal if repeated is True, the map will be rank 1.
- mud_examples.linear.models.createRandomLinearProblem(reference_point, num_qoi, num_observations, std_list, dist='normal', repeated=False)[source]#
Wrapper around createRandomLinearQoI to generalize to multiple QoI maps.
- mud_examples.linear.models.randA_gauss(dim_output, dim_input=None, seed=None)[source]#
Generate random Gaussian matrix, perform QR, and returns the resulting (orthogonal) Q
- mud_examples.linear.models.randA_list_svd(dim_output, dim_input=None, seed=None) List [source]#
Generate random square Gaussian matrix, perform SVD, and construct rank-1 matrices from components. Return list of them. Sum R entries of this returned list to generate a rank-R matrix.
- mud_examples.linear.models.randA_outer(dim_output, dim_input=None, seed=None)[source]#
Generate dimension rank-1 matrices using Gaussian entries to generate a vector x and then take outer-product with self.