familiarCollection-class: Collection of familiar data.

In familiar: End-to-End Automated Machine Learning and Model Evaluation

Collection of familiar data.

Description

A familiarCollection object aggregates data from one or more familiarData objects.

Slots

name

Name of the collection.

data_sets

Name of the individual underlying datasets.

outcome_type

Outcome type for which the collection was created.

outcome_info

Outcome information object, which contains information concerning the outcome, such as class levels.

fs_vimp

Variable importance data collected by feature selection methods.

model_vimp

Variable importance data collected from model-specific algorithms implemented by models created by familiar.

permutation_vimp

Data collected for permutation variable importance.

hyperparameters

Hyperparameters collected from created models.

hyperparameter_data

Additional data concerning hyperparameters. This is currently not used yet.

required_features

The set of features required for complete reproduction, i.e. with imputation.

model_features

The set of features that are required for using the model, but without imputation.

learner

Learning algorithm(s) used for data in the collection.

fs_method

Feature selection method(s) used for data in the collection.

prediction_data

Model predictions for the data in the collection.

confusion_matrix

Confusion matrix information for the data in the collection.

decision_curve_data

Decision curve analysis data for the data in the collection.

calibration_info

Calibration information, e.g. baseline survival in the development cohort.

calibration_data

Model calibration data collected from data in the collection.

model_performance

Collection of model performance data for data in the collection.

km_info

Information concerning risk-stratification cut-off values for data in the collection.

km_data

Kaplan-Meier survival data for data in the collection.

auc_data

AUC-ROC and AUC-PR data for data in the collection.

ice_data

Individual conditional expectation data for data in the collection. Partial dependence data are computed on the fly from these data.

univariate_analysis

Univariate analysis results of data in the collection.

feature_expressions

Feature expression values for data in the collection.

feature_similarity

Feature similarity information for data in the collection.

sample_similarity

Sample similarity information for data in the collection.

data_set_labels

Labels for the different datasets in the collection. See get_data_set_names and set_data_set_names.

learner_labels

Labels for the different learning algorithms used to create the collection. See get_learner_names and set_learner_names.

fs_method_labels

Labels for the different feature selection methods used to create the collection. See get_fs_method_names and set_fs_method_names.

feature_labels

Labels for the features in this collection. See get_feature_names and set_feature_names.

km_group_labels

Labels for the risk strata in this collection. See get_risk_group_names and set_risk_group_names.

class_labels

Labels of the response variable. See get_class_names and set_class_names.

project_id

Identifier of the project that generated this collection.

familiar_version

Version of the familiar package.

familiarCollection objects collect data from one or more familiarData objects. This objects are important, as all plotting and export functions use it. The fact that one can supply familiarModel, familiarEnsemble and familiarData objects as arguments for these methods, is because familiar internally converts these into familiarCollection objects prior to executing the method.

familiar documentation built on Sept. 30, 2024, 9:18 a.m.