Tools for Healthcare Machine Learning

---

Documentation for package ‘healthcareai’ version 2.5.1

• DESCRIPTION file.
• Package NEWS.

Help Pages

add_best_levels	Build efficient features from high-cardinality, multiple-membership factors
add_SAM_utility_cols	Add SAM utility columns to table
as.model_list	Make models into model_list object
build_connection_string	Build a connection string for use with MSSQL and dbConnect
catalyst_test_deploy_in_prod	Defunct
control_chart	Create a control chart
convert_date_cols	Convert character date columns to dates and times
db_read	Read from a SQL Server database table
evaluate	Get model performance metrics
evaluate.model_list	Get model performance metrics
evaluate.predicted_df	Get model performance metrics
evaluate_classification	Get performance metrics for classification predictions
evaluate_multiclass	Get performance metrics for multiclass predictions
evaluate_regression	Get performance metrics for regression predictions
explore	Explore a model's "reasoning" via counterfactual predictions
flash_models	Train models without tuning for performance
get_best_levels	Build efficient features from high-cardinality, multiple-membership factors
get_cutoffs	Get cutoff values for group predictions
get_hyperparameter_defaults	Get hyperparameter values
get_random_hyperparameters	Get hyperparameter values
get_supported_models	Supported models and their hyperparameters
get_thresholds	Get class-separating thresholds for classification predictions
get_variable_importance	Get variable importances
hcai_impute	Specify imputation methods for an existing recipe
healthcareai	Machine Learning Made Easy
hyperparameters	Get hyperparameter values
impute	Impute data and return a reusable recipe
interpret	Interpret a model via regularized coefficient estimates
is.classification_list	Type checks
is.model_list	Type checks
is.multiclass_list	Type checks
is.predicted_df	Class check
is.regression_list	Type checks
load_models	Save models to disk and load models from disk
machine_learn	Machine learning made easy
make_na	Replace missingness values with NA and correct columns types
missingness	Find missingness in each column and search for strings that might represent missing values
Mode	Mode
models	Supported models and their hyperparameters
models_supported	Supported models and their hyperparameters
pima_diabetes	Patient diabetes dataset
pima_meds	Patient medications dataset
pip	Patient Impact Predictor
pivot	Pivot multiple rows per observation to one row with multiple columns
plot.explore_df	Plot Counterfactual Predictions
plot.interpret	Plot regularized model coefficients
plot.missingness	Plot missingness
plot.model_list	Plot performance of models
plot.predicted_df	Plot model predictions vs observed outcomes
plot.thresholds_df	Plot threshold performance metrics
plot.variable_importance	Plot variable importance
plot_classification_predictions	Plot model predictions vs observed outcomes
plot_multiclass_predictions	Plot model predictions vs observed outcomes
plot_regression_predictions	Plot model predictions vs observed outcomes
predict.model_list	Get predictions
prep_data	Prepare data for machine learning
rename_with_counts	Adds the category count to each category name in a given variable column
save_models	Save models to disk and load models from disk
separate_drgs	Convert MSDRGs into a "base DRG" and complication level
split_train_test	Split data into training and test data frames
start_prod_logs	Defunct
step_add_levels	Add levels to nominal variables
step_date_hcai	Date and Time Feature Generator
step_dummy_hcai	Dummy Variables Creation
step_locfimpute	Last Observation Carried Forward Imputation
step_missing	Clean NA values from categorical/nominal variables
stop_prod_logs	Defunct
summary.missingness	Summarizes data given by 'missingness'
supported_models	Supported models and their hyperparameters
tidy.step_add_levels	Add levels to nominal variables
tidy.step_date_hcai	Date and Time Feature Generator
tidy.step_dummy_hcai	Dummy Variables Creation
tidy.step_locfimpute	Last Observation Carried Forward Imputation
tidy.step_missing	Clean NA values from categorical/nominal variables
tune_models	Tune multiple machine learning models using cross validation to optimize performance