Learn R Programming

iai: Interpretable AI R Interface

iai is a package providing an interface to the algorithms of Interpretable AI from the R programming language, including:

Optimal Trees for classification, regression, prescription and survival analysis
Optimal Imputation for missing data imputation and outlier detection
Optimal Feature Selection for exact sparse regression

Installation and Usage

Please refer to the official Interpretable AI documentation for information on setting up and using the package.

Copy Link

Version

Install

install.packages('iai')

Monthly Downloads

436

Version

1.10.2

License

MIT + file LICENSE

Maintainer

Jack Dunn

Last Published

October 18th, 2024

Functions in iai (1.10.2)

delete_rich_output_param

Delete a global rich output parameter

categorical_regression_reward_estimator

Learner for conducting reward estimation with categorical treatments and regression outcomes

Return an unfitted copy of a learner with the same parameters

fit.grid_search

Fits a grid_search to the training data

get_classification_label.classification_tree_multi_learner

Return the predicted label at a node of a multi-task tree

equal_propensity_estimator

Learner that estimates equal propensity for all treatments.

get_classification_proba

Generic function for returning the probabilities of class membership at a node of a classification tree

Generic function for fitting a learner.

fit_transform_cv

Train a grid using cross-validation with features and impute all missing values in these features

get_best_params

Return the best parameter combination from a grid

Fit an imputation learner with training features and create adaptive indicator features to encode the missing pattern

get_classification_proba.classification_tree_multi_learner

Return the predicted probabilities of class membership at a node of a multi-task tree

get_cluster_assignments

Return the indices of the trees assigned to each cluster, under the clustering of a given number of trees

get_classification_proba.classification_tree_learner

Return the predicted probabilities of class membership at a node of a tree

Fits a grid search to the training data with cross-validation

fit.imputation_learner

Fits an imputation learner to the training data.

get_cluster_distances

Return the distances between the centroids of each pair of clusters, under the clustering of a given number of trees

Get the depth of a node of a tree

Fits a model to the training data

get_cluster_details

Return the centroid information for each cluster, under the clustering of a given number of trees

Return the machine ID for the current computer.

get_num_samples

Get the number of training points contained in a node of a tree

get_lower_child

Get the index of the lower child at a split node of a tree

Return the value of all parameters on a learner

get_prescription_treatment_rank

Return the treatments ordered from most effective to least effective at a node of a tree

get_classification_label

Generic function for returning the predicted label in the node of a classification tree

fit.optimal_feature_selection_learner

Fits an Optimal Feature Selection learner to the training data

get_regression_constant.classification_tree_learner

Return the constant term in the logistic regression prediction at a node of a classification tree

get_roc_curve_data

Extract the underlying data from an ROC curve

get_split_categories

Return the categoric/ordinal information used in the split at a node of a tree

get_regression_constant.classification_tree_multi_learner

Return the constant term in the logistic regression prediction at a node of a multi-task classification tree

get_grid_results

Return a summary of the results from the grid search

Generic function for fitting a reward estimator on features, treatments and returning predicted counterfactual rewards and scores of the internal estimators.

Return the fitted learner using the best parameter combination from a grid

fit_predict.categorical_reward_estimator

Fit a categorical reward estimator on features, treatments and outcomes and return predicted counterfactual rewards for each observation, under each treatment observed in the data, as well as the scores of the internal estimators.

is_hyperplane_split

Check if a node of a tree applies a hyperplane split

get_classification_label.classification_tree_learner

Return the predicted label at a node of a tree

get_train_errors

Extract the training objective value for each candidate tree in the comparison, where a lower value indicates a better solution

Return a copy of the learner that uses a specific tree rather than the tree with the best training objective.

get_estimation_densities

Return the total kernel density surrounding each treatment candidate for the propensity/outcome estimation problems in a fitted learner.

Check if a node of a tree is a leaf

fit_predict.numeric_reward_estimator

Fit a numeric reward estimator on features, treatments and outcomes and return predicted counterfactual rewards for each observation, under each treatment candidate, as well as the scores of the internal estimators.

is_ordinal_split

Check if a node of a tree applies a ordinal split

is_parallel_split

Check if a node of a tree applies a parallel split

get_features_used

Return the names of the features used by the learner

multi_tree_plot

Generic function for constructing an interactive tree visualization of multiple tree learners

Generic function for returning the number of fits in a trained learner

get_num_fits.glmnetcv_learner

Return the number of fits along the path in a trained GLMNet learner

get_grid_result_summary

Return a summary of the results from the grid search

get_grid_result_details

Return a vector of lists detailing the results of the grid search

multi_tree_plot.default

Construct an interactive tree visualization of multiple tree learners as specified by questions

optimal_feature_selection_regressor

Learner for conducting Optimal Feature Selection on regression problems

optimal_feature_selection_classifier

Learner for conducting Optimal Feature Selection on classification problems

get_prediction_weights.optimal_feature_selection_learner

Return the weights for numeric and categoric features used for prediction in a trained Optimal Feature Selection learner

get_prediction_weights.glmnetcv_learner

Return the weights for numeric and categoric features used for prediction in a trained GLMNet learner

Fit an imputation model using the given features and impute the missing values in these features

get_rich_output_params

Return the current global rich output parameter settings

get_regression_weights.survival_tree_learner

Return the weights for each feature in the cox regression prediction at a node of a survival tree

get_regression_constant.survival_tree_learner

Return the constant term in the cox regression prediction at a node of a survival tree

get_regression_constant.regression_tree_multi_learner

Return the constant term in the linear regression prediction at a node of a multi-task regression tree

get_num_fits.optimal_feature_selection_learner

Return the number of fits along the path in a trained Optimal Feature Selection learner

Return the number of nodes in a trained learner

predict.categorical_reward_estimator

Return counterfactual rewards estimated by a categorical reward estimator for each observation in the supplied data

Generic function for returning the predictions of a model

get_regression_constant

Generic function for returning the constant term in the regression prediction at a node of a tree

get_policy_treatment_outcome_standard_error

Return the standard error for the quality of the treatments at a node of a tree

get_policy_treatment_rank

Return the treatments ordered from most effective to least effective at a node of a tree

get_regression_constant.prescription_tree_learner

Return the constant term in the linear regression prediction at a node of a prescription tree

get_regression_constant.regression_tree_learner

Return the constant term in the linear regression prediction at a node of a regression tree

get_regression_weights.regression_tree_learner

Return the weights for each feature in the linear regression prediction at a node of a regression tree

Get the index of the parent node at a node of a tree

get_policy_treatment_outcome

Return the quality of the treatments at a node of a tree

get_regression_weights.regression_tree_multi_learner

Return the weights for each feature in the linear regression prediction at a node of a multi-task regression tree

get_prediction_constant.optimal_feature_selection_learner

Return the constant term in the prediction in a trained Optimal Feature Selection learner

get_regression_weights

Generic function for returning the weights for each feature in the regression prediction at a node of a tree

glmnetcv_regressor

Learner for training GLMNet models for regression problems with cross-validation

get_upper_child

Get the index of the upper child at a split node of a tree

glmnetcv_classifier

Learner for training GLMNet models for classification problems with cross-validation

get_prediction_weights

Generic function for returning the prediction weights in a trained learner

Generic function for returning the hazard coefficient predicted by a model

glmnetcv_survival_learner

Learner for training GLMNet models for survival problems with cross-validation

get_regression_weights.classification_tree_multi_learner

Return the weights for each feature in the logistic regression prediction at a node of a multi-task classification tree

Download and install Julia automatically.

Impute missing values using cross validation

get_split_weights

Return the weights for numeric and categoric features used in the hyperplane split at a node of a tree

opt_knn_imputation_learner

Learner for conducting optimal k-NN imputation

get_stability_results

Return the trained trees in order of increasing objective value, along with their variable importance scores for each feature

get_prediction_constant.glmnetcv_learner

Return the constant term in the prediction in a trained GLMNet learner

get_prediction_constant

Generic function for returning the prediction constant in a trained learner

numeric_survival_reward_estimator

Learner for conducting reward estimation with numeric treatments and survival outcomes

Controls grid search over parameter combinations

get_survival_curve

Return the survival curve at a node of a tree

get_regression_weights.prescription_tree_learner

Return the weights for each feature in the linear regression prediction at a node of a prescription tree

get_survival_curve_data

Extract the underlying data from a survival curve (as returned by predict.survival_learner or get_survival_curve)

predict_outcomes.policy_learner

Return the predicted outcome for each treatment made by a policy learner for each point in the features

predict_hazard.glmnetcv_survival_learner

Return the fitted hazard coefficient estimate made by a glmnetcv_survival_learner for each point in the features.

predict_outcomes.prescription_learner

Return the predicted outcome for each treatment made by a prescription learner for each point in the features

multi_questionnaire

Generic function for constructing an interactive questionnaire with multiple learners

Initialize Julia and the IAI package.

opt_svm_imputation_learner

Learner for conducting optimal SVM imputation

missing_goes_lower

Check if points with missing values go to the lower child at a split node of of a tree

get_regression_weights.classification_tree_learner

Return the weights for each feature in the logistic regression prediction at a node of a classification tree

multi_questionnaire.default

Construct an interactive questionnaire from multiple specified learners

install_system_image

Download and install the IAI system image automatically.

numeric_reward_estimator

Learner for conducting reward estimation with numeric treatments

multi_questionnaire.grid_search

Construct an interactive tree questionnaire using multiple learners from the results of a grid search

opt_tree_imputation_learner

Learner for conducting optimal tree-based imputation

plot.similarity_comparison

Plot a similarity comparison

predict_expected_survival_time.survival_curve

Return the expected survival time estimate made by a survival curve (as returned by predict.survival_learner or get_survival_curve)

optimal_tree_prescription_maximizer

Learner for training Optimal Prescriptive Trees where the prescriptions should aim to maximize outcomes

numeric_regression_reward_estimator

Learner for conducting reward estimation with numeric treatments and regression outcomes

plot.stability_analysis

Plot a stability analysis

predict_expected_survival_time

Generic function for returning the expected survival time predicted by a model

is_categoric_split

Check if a node of a tree applies a categoric split

optimal_tree_policy_minimizer

Learner for training Optimal Policy Trees where the policy should aim to minimize outcomes

Generic function for returning the probabilities of class membership predicted by a model

questionnaire.optimal_feature_selection_learner

Specify an interactive questionnaire of an Optimal Feature Selection learner

optimal_tree_classifier

Learner for training Optimal Classification Trees

predict_expected_survival_time.survival_learner

Return the expected survival time estimate made by a survival learner for each point in the features.

predict_expected_survival_time.glmnetcv_survival_learner

Return the expected survival time estimate made by a glmnetcv_survival_learner for each point in the features.

questionnaire.tree_learner

Specify an interactive questionnaire of a tree learner

get_split_feature

Return the feature used in the split at a node of a tree

imputation_learner

Generic learner for imputing missing values

get_survival_hazard

Return the predicted hazard ratio at a node of a tree

is_mixed_ordinal_split

Check if a node of a tree applies a mixed ordinal/categoric split

get_survival_expected_time

Return the predicted expected survival time at a node of a tree

Impute missing values using either a specified method or through validation

predict_treatment_outcome

Return the estimated quality of each treatment in the trained model of the learner for each point in the features

predict_treatment_outcome_standard_error

Return the standard error for the estimated quality of each treatment in the trained model of the learner for each point in the features

resume_from_checkpoint

Resume training from a checkpoint file

reward_estimator

Learner for conducting reward estimation with categorical treatments

get_split_threshold

Return the threshold used in the split at a node of a tree

mean_imputation_learner

Learner for conducting mean imputation

Loads the Julia Graphviz library to permit certain visualizations.

optimal_tree_survival_learner

Learner for training Optimal Survival Trees

optimal_tree_survivor

Learner for training Optimal Survival Trees

optimal_tree_multi_classifier

Learner for training multi-task Optimal Classification Trees

predict.optimal_feature_selection_learner

Return the predictions made by an Optimal Feature Selection learner for each point in the features

numeric_classification_reward_estimator

Learner for conducting reward estimation with numeric treatments and classification outcomes

multi_tree_plot.grid_search

Construct an interactive tree visualization of multiple tree learners from the results of a grid search

predict_proba.classification_learner

Return the probabilities of class membership predicted by a classification learner for each point in the features

predict_treatment_rank

Return the treatments in ranked order of effectiveness for each point in the features

release_license

Release any IAI license held by the current session.

random_forest_regressor

Learner for training random forests for regression problems

is_mixed_parallel_split

Check if a node of a tree applies a mixed parallel/categoric split

random_forest_survival_learner

Learner for training random forests for survival problems

Generic function for constructing an ROC curve

Print the decision path through the learner for each sample in the features

score.glmnetcv_learner

Calculate the score for a GLMNet learner on the given data

show_in_browser.tree_learner

Show interactive tree visualization of a tree learner in the default browser

score.numeric_reward_estimator

Calculate the scores for a numeric reward estimator on the given data

show_questionnaire

Generic function for showing interactive questionnaire in browser

variable_importance.learner

Generate a ranking of the variables in a learner according to their importance during training. The results are normalized so that they sum to one.

variable_importance

Generic function for calculating variable importance

optimal_tree_policy_maximizer

Learner for training Optimal Policy Trees where the policy should aim to maximize outcomes

optimal_tree_multi_regressor

Learner for training multi-task Optimal Regression Trees

Calculate SHAP values for all points in the features using the learner

predict_hazard.survival_learner

Return the fitted hazard coefficient estimate made by a survival learner for each point in the features.

predict_outcomes

Generic function for returning the outcomes predicted by a model under each treatment

predict_reward.numeric_reward_estimator

Return counterfactual rewards estimated by a numeric reward estimator for each observation in the supplied data and predictions

predict.supervised_learner

Return the predictions made by a supervised learner for each point in the features

Output a learner in .dot format

Read in a learner or grid saved in JSON format

optimal_tree_prescription_minimizer

Learner for training Optimal Prescriptive Trees where the prescriptions should aim to minimize outcomes

predict_proba.classification_multi_learner

Return the probabilities of class membership predicted by a multi-task classification learner for each point in the features

plot.grid_search

Plot a grid search results for Optimal Feature Selection learners

Plot an ROC curve

predict_proba.glmnetcv_classifier

Return the probabilities of class membership predicted by a glmnetcv_classifier learner for each point in the features

reset_display_label

Reset the predicted probability displayed to be that of the predicted label when visualizing a learner

Calculate the score for a set of predictions on the given data

score.categorical_reward_estimator

Calculate the scores for a categorical reward estimator on the given data

roc_curve.classification_learner

Construct an ROC curve using a trained classification learner on the given data

set_rich_output_param

Sets a global rich output parameter

set_reward_kernel_bandwidth

Save a new reward kernel bandwidth inside a learner, and return new reward predictions generated using this bandwidth for the original data used to train the learner.

predict.glmnetcv_learner

Return the predictions made by a GLMNet learner for each point in the features

optimal_tree_regressor

Learner for training Optimal Regression Trees

predict.numeric_reward_estimator

Return counterfactual rewards estimated by a numeric reward estimator for each observation in the supplied data

Generic function for writing interactive visualization to file

predict.supervised_multi_learner

Return the predictions made by a multi-task supervised learner for each point in the features

Generic function for constructing an interactive questionnaire

predict_reward.categorical_reward_estimator

Return counterfactual rewards estimated by a categorical reward estimator for each observation in the supplied data and predictions

predict.survival_learner

Return the predictions made by a survival learner for each point in the features

Use the trained trees in a learner along with the supplied validation data to determine the best value for the `cp` parameter and then prune the trees according to this value

Generic function for returning the counterfactual rewards estimated by a model under each treatment

roc_curve.glmnetcv_classifier

Construct an ROC curve using a trained glmnetcv_classifier on the given data

Refit the models in the leaves of a trained learner using the supplied data

show_in_browser.abstract_visualization

Show interactive visualization of an object in the default browser

Generic function for calculating scores

write_html.abstract_visualization

Output an object as an interactive browser visualization in HTML format

score.supervised_multi_learner

Calculate the score for a multi-task model on the given data

variable_importance_similarity

Calculate similarity between the final tree in a tree learner with all trees in new tree learner using variable importance scores.

Write the internal booster saved in the learner to file

set_display_label

Show the probability of a specified label when visualizing a learner

Split the data into training and test datasets

show_in_browser

Generic function for showing interactive visualization in browser

stability_analysis

Conduct a stability analysis of the trees in a tree learner

For a binary classification problem, update the the predicted labels in the leaves of the learner to predict a label only if the predicted probability is at least the specified threshold.

Impute missing values in a dataframe using a fitted imputation model

write_html.roc_curve

Output an ROC curve as an interactive browser visualization in HTML format

write_questionnaire.tree_learner

Output a tree learner as an interactive questionnaire in HTML format

Output a learner or grid in JSON format

show_in_browser.roc_curve

Show interactive visualization of a roc_curve in the default browser

write_html.tree_learner

Output a tree learner as an interactive browser visualization in HTML format

Output a learner as a SVG image

similarity_comparison

Conduct a similarity comparison between the final tree in a learner and all trees in a new learner to consider the tradeoff between training performance and similarity to the original tree

score.supervised_learner

Calculate the score for a model on the given data

score.optimal_feature_selection_learner

Calculate the score for an Optimal Feature Selection learner on the given data

roc_curve.default

Construct an ROC curve from predicted probabilities and true labels

random_forest_classifier

Learner for training random forests for classification problems

roc_curve.classification_multi_learner

Construct an ROC curve using a trained multi-task classification learner on the given data

Set the random seed in Julia

Set all supplied parameters on a learner

rand_imputation_learner

Learner for conducting random imputation

transform_and_expand

Transform features with a trained imputation learner and create adaptive indicator features to encode the missing pattern

Output a learner as a PNG image

Output a learner as a PDF image

xgboost_survival_learner

Learner for training XGBoost models for survival problems

zero_imputation_learner

Learner for conducting zero-imputation

single_knn_imputation_learner

Learner for conducting heuristic k-NN imputation

write_questionnaire

Generic function for writing interactive questionnaire to file

variable_importance.tree_learner

Generate a ranking of the variables in a tree learner according to their importance during training. The results are normalized so that they sum to one.

variable_importance.optimal_feature_selection_learner

Generate a ranking of the variables in an Optimal Feature Selection learner according to their importance during training. The results are normalized so that they sum to one.

write_questionnaire.optimal_feature_selection_learner

Output an Optimal Feature Selection learner as an interactive questionnaire in HTML format

show_questionnaire.tree_learner

Show an interactive questionnaire based on a tree learner in default browser

xgboost_classifier

Learner for training XGBoost models for classification problems

Specify an interactive tree visualization of a tree learner

show_questionnaire.optimal_feature_selection_learner

Show an interactive questionnaire based on an Optimal Feature Selection learner in default browser

xgboost_regressor

Learner for training XGBoost models for regression problems

tune_reward_kernel_bandwidth

Conduct the reward kernel bandwidth tuning procedure for a range of starting bandwidths and return the final tuned values.

autoplot.roc_curve

Construct a ggplot2::ggplot object plotting the ROC curve

add_julia_processes

Add additional Julia worker processes to parallelize workloads

Return the indices of the points in the features that fall into each node of a trained tree model

acquire_license

Acquire an IAI license for the current session.

autoplot.grid_search

Construct a ggplot2::ggplot object plotting grid search results for Optimal Feature Selection learners

autoplot.similarity_comparison

Construct a ggplot2::ggplot object plotting the results of the similarity comparison

autoplot.stability_analysis

Construct a ggplot2::ggplot object plotting the results of the stability analysis

all_treatment_combinations

Return a dataframe containing all treatment combinations of one or more treatment vectors, ready for use as treatment candidates in `fit_predict!` or `predict`

Return the leaf index in a tree model into which each point in the features falls

Convert a vector of values to IAI mixed data format

cleanup_installation

Remove all traces of automatic Julia/IAI installation

convert_treatments_to_numeric

Convert `treatments` from symbol/string format into numeric values.

copy_splits_and_refit_leaves

Copy the tree split structure from one learner into another and refit the models in each leaf of the tree using the supplied data

categorical_classification_reward_estimator

Learner for conducting reward estimation with categorical treatments and classification outcomes

categorical_reward_estimator

Learner for conducting reward estimation with categorical treatments

categorical_survival_reward_estimator

Learner for conducting reward estimation with categorical treatments and survival outcomes

Return a matrix where entry (i, j) is true if the ith point in the features passes through the jth node in a trained tree model.