API

List of functions and classes in Himalaya.

Backend

Public functions in himalaya.backend.

set_backend	Set the backend using a global variable, and return the backend module.
get_backend	Get the current backend module.
ALL_BACKENDS	Built-in mutable sequence.

---

Kernel ridge

Public functions and classes in himalaya.kernel_ridge.

Estimators

Estimators compatible with the scikit-learn API.

KernelRidge	Kernel ridge regression.
KernelRidgeCV	Kernel ridge regression with efficient cross-validation over alpha.
WeightedKernelRidge	Weighted kernel ridge regression.
MultipleKernelRidgeCV	Multiple-kernel ridge regression with cross-validation.
Kernelizer	Transform tabular data into a kernel.
ColumnKernelizer	Applies transformers to columns of an array, ending with kernelizers.
make_column_kernelizer	Construct a ColumnKernelizer from the given transformers.

Solver functions

KERNEL_RIDGE_SOLVERS
solve_kernel_ridge_cv_eigenvalues	Solve kernel ridge regression with a grid search over alphas.
solve_kernel_ridge_eigenvalues	Solve kernel ridge regression using eigenvalues decomposition.
solve_kernel_ridge_gradient_descent	Solve kernel ridge regression using conjugate gradient.
solve_kernel_ridge_conjugate_gradient	Solve kernel ridge regression using conjugate gradient.
WEIGHTED_KERNEL_RIDGE_SOLVERS
solve_weighted_kernel_ridge_gradient_descent	Solve weighted kernel ridge regression using gradient descent.
solve_weighted_kernel_ridge_conjugate_gradient	Solve weighted kernel ridge regression using conjugate gradient.
solve_weighted_kernel_ridge_neumann_series	Solve weighted kernel ridge regression using Neumann series.
MULTIPLE_KERNEL_RIDGE_SOLVERS
solve_multiple_kernel_ridge_hyper_gradient	Solve bilinear kernel ridge regression with cross-validation.
solve_multiple_kernel_ridge_random_search	Solve multiple kernel ridge regression using random search.

Helpers

generate_dirichlet_samples	Generate samples from a Dirichlet distribution.
predict_weighted_kernel_ridge	Compute predictions, typically on a test set.
predict_and_score_weighted_kernel_ridge	Compute predictions, typically on a test set, and compute the score.
primal_weights_kernel_ridge	Compute the primal weights for kernel ridge regression.
primal_weights_weighted_kernel_ridge	Compute the primal weights for weighted kernel ridge regression.

Kernels

PAIRWISE_KERNEL_FUNCTIONS
linear_kernel	Compute the linear kernel between X and Y.
polynomial_kernel	Compute the polynomial kernel between X and Y.
rbf_kernel	Compute the rbf (gaussian) kernel between X and Y.
sigmoid_kernel	Compute the sigmoid kernel between X and Y.
cosine_similarity_kernel	Compute cosine similarity between samples in X and Y.
KernelCenterer	Center a kernel matrix.

---

Lasso

Public functions and classes in himalaya.lasso.

Estimators

Estimators compatible with the scikit-learn API.

SparseGroupLassoCV

Sparse group Lasso

Solver functions

solve_sparse_group_lasso	Solves the sparse group Lasso.
solve_sparse_group_lasso_cv	Solves the sparse group Lasso, selecting hyperparameters over cross-validation.

---

Ridge

Public functions and classes in himalaya.ridge.

Estimators

Estimators compatible with the scikit-learn API.

Ridge	Ridge regression.
RidgeCV	Ridge regression with efficient cross-validation over alpha.
GroupRidgeCV	Group ridge regression with cross-validation.
BandedRidgeCV	alias of himalaya.ridge._sklearn_api.GroupRidgeCV
ColumnTransformerNoStack	Applies transformers to columns of an array, and does not stack them.
make_column_transformer_no_stack	Construct a ColumnTransformerNoStack from the given transformers.

Solver functions

RIDGE_SOLVERS
solve_ridge_svd	Solve ridge regression using SVD decomposition.
solve_ridge_cv_svd	Solve ridge regression with a grid search over alphas.
GROUP_RIDGE_SOLVERS
BANDED_RIDGE_SOLVERS
solve_group_ridge_random_search	Solve group ridge regression using random search on the simplex.
solve_banded_ridge_random_search	Solve group ridge regression using random search on the simplex.

---

Other modules

Public functions and classes in other minor modules.

Progress bar

progress_bar.ProgressBar	Generate a command-line progress bar.
progress_bar.bar	Simple API for progress_bar.

Scoring functions

scoring.l2_neg_loss	L2 negative loss, computed for multiple predictions.
scoring.r2_score	R2 score, computed for multiple predictions (e.g.
scoring.correlation_score	Correlation score, computed for multiple predictions.
scoring.r2_score_split	Split the R2 score into individual components using the product measure.
scoring.r2_score_split_svd	Split the R2 score into individual components using relative weights.
scoring.correlation_score_split	Split the correlation score into individual components.

Utils

utils.compute_lipschitz_constants	Compute Lipschitz constants of gradients of linear regression problems.
utils.generate_multikernel_dataset	Utility to generate datasets for the gallery of examples.

Visualization

viz.plot_alphas_diagnostic

Plot a diagnostic plot for the selected alphas during cross-validation.