nonconformist API¶
nc¶
Nonconformity functions.
Nonconformity Functions¶
nc.BaseModelNc(model, err_func[, …]) |
Base class for nonconformity scorers based on an underlying model. |
nc.ClassifierNc(model[, err_func, …]) |
Nonconformity scorer using an underlying class probability estimating model. |
nc.RegressorNc(model[, err_func, …]) |
Nonconformity scorer using an underlying regression model. |
Error Functions¶
nc.ClassificationErrFunc() |
Base class for classification model error functions. |
nc.RegressionErrFunc() |
Base class for regression model error functions. |
nc.InverseProbabilityErrFunc() |
Calculates the probability of not predicting the correct class. |
nc.MarginErrFunc() |
Calculates the margin error. |
nc.AbsErrorErrFunc() |
Calculates absolute error nonconformity for regression problems. |
nc.SignErrorErrFunc() |
Calculates signed error nonconformity for regression problems. |
icp¶
Inductive conformal predictors.
Classes¶
icp.IcpClassifier(nc_function[, condition, …]) |
Inductive conformal classifier. |
icp.IcpRegressor(nc_function[, condition]) |
Inductive conformal regressor. |
acp¶
Aggregated conformal predictors
Classes¶
acp.AggregatedCp(predictor[, sampler, …]) |
Aggregated conformal predictor. |
acp.RandomSubSampler([calibration_portion]) |
Random subsample sampler. |
acp.BootstrapSampler |
Bootstrap sampler. |
acp.CrossSampler |
Cross-fold sampler. |
evaluation¶
Evaluation of conformal predictors.
Classes¶
evaluation.ClassIcpCvHelper(icp[, …]) |
Helper class for running the cross_val_score evaluation method on IcpClassifiers. |
evaluation.RegIcpCvHelper(icp[, …]) |
Helper class for running the cross_val_score evaluation method on IcpRegressors. |
Functions¶
evaluation.cross_val_score(model, x, y[, …]) |
Evaluates a conformal predictor using cross-validation. |
evaluation.run_experiment(models, csv_files) |
Performs a cross-validation evaluation of one or several conformal predictors on a collection of data sets in csv format. |
evaluation.reg_mean_errors(prediction, y, …) |
Calculates the average error rate of a conformal regression model. |
evaluation.class_mean_errors(prediction, y) |
Calculates the average error rate of a conformal classification model. |
evaluation.reg_mean_size(prediction, y, …) |
Calculates the average prediction interval size of a conformal regression model. |
evaluation.class_avg_c(prediction, y, …) |
Calculates the average number of classes per prediction of a conformal classification model. |
evaluation.class_one_c(prediction, y, …) |
Calculates the rate of singleton predictions (prediction sets containing only a single class label) of a conformal classification model. |
evaluation.class_mean_p_val(prediction, y, …) |
Calculates the mean of the p-values output by a conformal classification model. |
