Module functions 

Auto-generated module

🤖 Generated with SplitRS

Functions

accuracy_from_cm

Compute overall accuracy from a confusion matrix.

auc_trapezoid

Compute AUC using the trapezoidal rule.

average_path_length

Expected path length of a random BST with n nodes.

backward_substitution_t

Backward substitution against Lᵀ: solve Lᵀ x = b.

best_split

Find the best split for a set of samples.

build_isolation_tree

Build one isolation tree from a subsample.

build_tree

cholesky

Cholesky decomposition (lower triangular factor).

compute_pairwise_affinities

Compute pairwise affinities p_{ij} (symmetric, normalised) for t-SNE.

confusion_matrix

Compute the confusion matrix for binary or multi-class classification.

cross_val_linear_regression

Run k-fold cross-validation for a linear regression model (R² metric).

dot

feature_bagging

Select a random subset of k feature indices from 0..d.

forward_substitution

Forward substitution: solve L x = b where L is lower-triangular.

gini_impurity

Compute Gini impurity of a label set.

kernel_matrix

Compute the kernel matrix K[i, j] = k(X[i], X[j]).

kmeans

Run k-means clustering with k-means++ initialisation.

lof_scores

Compute LOF scores for all points in x.

macro_f1

Macro-averaged F1 across all classes.

mae

Mean absolute error.

majority_vote

mean_vec

mse

Mean squared error.

multiclass_prf

Per-class precision, recall, and F1 for multi-class problems (one-vs-rest).

pca_fit

Fit PCA using the covariance matrix and power iteration.

power_iteration

Power iteration to find the dominant eigenpair of a symmetric matrix.

precision_recall_f1

Compute precision, recall, and F1 score from a binary confusion matrix.

r2

R² (coefficient of determination).

random_forest_oob

Compute the out-of-bag error for a random forest.

rbf_kernel

RBF (squared-exponential) kernel.

reach_dist

Compute the reachability distance: max(k-dist(o), dist(p, o)).

rmse

Root mean squared error.

roc_auc

Compute the ROC AUC (Area Under the Curve) for binary classification.

roc_curve

Compute the full ROC curve: returns (fpr, tpr) sorted by threshold.

sigmoid

sigmoid_vec

solve_linear_system

Solve Ax = b via Gaussian elimination with partial pivoting.

standardize

tsne

Run a simplified t-SNE reduction to 2 dimensions.

variance_vec

vec_add

vec_norm

vec_scale

vec_sub

weighted_f1

Weighted-averaged F1 (weighted by class support).