pertpy.tools.PerturbationComparison

class PerturbationComparison

Comparison between real and simulated perturbations.

Methods table

compare_classification(real, simulated, control)	Compare classification accuracy between real and simulated perturbations.
compare_knn(real, simulated[, control, ...])	Calculate proportions of real perturbed and control data points for simulated data.

Methods

PerturbationComparison.compare_classification(real, simulated, control, clf=None)

Compare classification accuracy between real and simulated perturbations.

Trains a classifier on the real perturbation data & the control data and reports a normalized classification accuracy on the simulated perturbation.

Parameters:

• real (ndarray) – Real perturbed data.

• simulated (ndarray) – Simulated perturbed data.

• control (ndarray) – Control data

• clf (ClassifierMixin | None, default: None) – sklearn classifier to use, sklearn.linear_model.LogisticRegression if not provided.

Return type:

float

PerturbationComparison.compare_knn(real, simulated, control=None, use_simulated_for_knn=False, n_neighbors=20, random_state=0, n_jobs=1)

Calculate proportions of real perturbed and control data points for simulated data.

Computes proportions of real perturbed, control and simulated (if use_simulated_for_knn=True) data points for simulated data. If control (C) is not provided, builds the knn graph from real perturbed + simulated perturbed.

Parameters:

• real (ndarray) – Real perturbed data.

• simulated (ndarray) – Simulated perturbed data.

• control (ndarray | None, default: None) – Control data

• use_simulated_for_knn (bool, default: False) – Include simulted perturbed data (simulated) into the knn graph. Only valid when control (control) is provided.

• n_neighbors (int, default: 20) – Number of neighbors to use in k-neighbor graph.

• random_state (int, default: 0) – Random state used for k-neighbor graph construction.

• n_jobs (int, default: 1) – Number of cores to use. Defaults to -1 (all).

Return type:

dict[str, float]