emil: Evaluation of Modeling without Information Leakage

A toolbox for designing and evaluating predictive models with resampling methods. The aim of this package is to provide a simple and efficient general framework for working with any type of prediction problem, be it classification, regression or survival analysis, that is easy to extend and adapt to your specific setting. Some commonly used methods for classification, regression and survival analysis are included.

AuthorChristofer Backlin [aut, cre], Mats Gustafsson [aut]
Date of publication2016-06-21 07:48:48
MaintainerChristofer Backlin <emil@christofer.backlin.se>
LicenseGPL (>= 2)

View on CRAN

Man pages

as.modeling_procedure: Coerce to modeling procedure

dichotomize: Dichotomize time-to-event data

emil: Introduction to the emil package

error_fun: Performance estimation functions

evaluate: Evaluate a modeling procedure

extension: Extending the emil framework with user-defined methods

factor_to_logical: Convert factors to logicals

fill: Replace values with something else

fit: Fit a model

fit_caret: Fit a model using the 'caret' package

fit_cforest: Fit conditional inference forest

fit_coxph: Fit Cox proportional hazards model

fit_glmnet: Fit elastic net, LASSO or ridge regression model

fit_lda: Fit linear discriminant

fit_lm: Fit a linear model fitted with ordinary least squares

fit_naive_bayes: Fit a naive Bayes classifier

fit_pamr: Fit nearest shrunken centroids model.

fit_qda: Fit quadratic discriminant.

fit_randomForest: Fit random forest.

fit_rpart: Fit a decision tree

fit_svm: Fit a support vector machine

get_color: Get color palettes

get_importance: Feature (variable) importance of a fitted model

get_performance: Extract prediction performance

get_prediction: Extract predictions from modeling results

get_response: Extract the response from a data set

get_tuning: Extract parameter tuning statistics

image.resample: Visualize resampling scheme

importance_glmnet: Feature importance extractor for elastic net models

importance_pamr: Feature importance of nearest shrunken centroids.

importance_randomForest: Feature importance of random forest.

impute: Regular imputation

indent: Increase indentation

index_fit: Convert a fold to row indexes of fittdng or test set

is_blank: Wrapper for several methods to test if a variable is empty

is_constant: Check if an object contains more than one unique value

is_multi_procedure: Detect if modeling results contains multiple procedures

learning_curve: Learning curve analysis

list_method: List all available methods

log_message: Print a timestamped and indented log message

mode: Get the most common value

modeling_procedure: Setup a modeling procedure

na_index: Support function for identifying missing values

name_procedure: Get names for modeling procedures

neg_gmpa: Negative geometric mean of class specific predictive accuracy

nice_axis: Plots an axis the way an axis should be plotted.

nice_box: Plots a box around a plot

nice_require: Load a package and offer to install if missing

notify_once: Print a warning message if not printed earlier

pipe: Pipe operator

plot.learning_curve: Plot results from learning curve analysis

plot.Surv: Plot Surv vector

predict_caret: Predict using a 'caret' method

predict_cforest: Predict with conditional inference forest

predict_coxph: Predict using Cox proportional hazards model

predict_glmnet: Predict using generalized linear model with elastic net...

predict_lda: Prediction using already trained prediction model

predict_lm: Prediction using linear model

predict.model: Predict the response of unknown observations

predict_naive_bayes: Predict using naive Bayes model

predict_pamr: Prediction using nearest shrunken centroids.

predict_qda: Prediction using already trained classifier.

predict_randomForest: Prediction using random forest.

predict_rpart: Predict using a fitted decision tree

predict_svm: Predict using support vector machine

pre_factor_to_logical: Convert factors to logical columns

pre_impute: Basic imputation

pre_impute_df: Impute a data frame

pre_impute_knn: Nearest neighbors imputation

pre_log_message: Print log message during pre-processing

pre_pamr: PAMR adapted dataset pre-processing

pre_process: Data preprocessing

print.preprocessed_data: Print method for pre-processed data

pvalue: Extraction of p-value from a statistical test

pvalue.coxph: Extract p-value from a Cox proportional hazards model

pvalue.crr: Extracts p-value from a competing risk model

pvalue.cuminc: Extract p-value from a cumulative incidence estimation

pvalue.survdiff: Extracts p-value from a logrank test

resample: Resampling schemes

roc_curve: Calculate ROC curves

select: 'emil' and 'dplyr' integration

subresample: Generate resampling subschemes

subtree: Extract a subset of a tree of nested lists

trivial_error_rate: Calculate the trivial error rate

tune: Tune parameters of modeling procedures

validate_data: Validate a pre-processed data set

vlines: Add vertical or horizontal lines to a plot

weighted_error_rate: Weighted error rate


\%>\% Man page
as.data.frame.roc_curve Man page
as.data.table.roc_curve Man page
as.modeling_procedure Man page
detune Man page
dichotomize Man page
emil Man page
error_fun Man page
error_rate Man page
evaluate Man page
extension Man page
factor_to_logical Man page
fill Man page
fit Man page
fit_caret Man page
fit_cforest Man page
fit_coxph Man page
fit_glmnet Man page
fit_lasso Man page
fit_lda Man page
fit_lm Man page
fit_naive_bayes Man page
fit_pamr Man page
fit_qda Man page
fit_randomForest Man page
fit_ridge_regression Man page
fit_rpart Man page
fit_svm Man page
get_color Man page
get_color.default Man page
get_color.factor Man page
get_importance Man page
get_performance Man page
get_prediction Man page
get_response Man page
get_tuning Man page
hlines Man page
image.crossvalidation Man page
image.resample Man page
importance_glmnet Man page
importance_lasso Man page
importance_pamr Man page
importance_randomForest Man page
importance_ridge_regression Man page
impute Man page
impute_knn Man page
impute_median Man page
indent Man page
index_fit Man page
index_test Man page
is_blank Man page
is_constant Man page
is_multi_procedure Man page
is_tunable Man page
is_tuned Man page
learning_curve Man page
list_method Man page
log_message Man page
mode Man page
modeling_procedure Man page
mse Man page
na_fill Man page
na_index Man page
name_procedure Man page
neg_auc Man page
neg_gmpa Man page
neg_harrell_c Man page
nice_axis Man page
nice_box Man page
nice_require Man page
notify_once Man page
plot.learning_curve Man page
plot.roc_curve Man page
plot.Surv Man page
pre_center Man page
pre_convert Man page
predict_caret Man page
predict_cforest Man page
predict_coxph Man page
predict_glmnet Man page
predict_lasso Man page
predict_lda Man page
predict_lm Man page
predict.model Man page
predict_naive_bayes Man page
predict_pamr Man page
predict_qda Man page
predict_randomForest Man page
predict_ridge_regression Man page
predict_rpart Man page
predict_svm Man page
pre_factor_to_logical Man page
pre_impute Man page
pre_impute_df Man page
pre_impute_knn Man page
pre_impute_mean Man page
pre_impute_median Man page
pre_log_message Man page
pre_pamr Man page
pre_pca Man page
pre_process Man page
pre_remove Man page
pre_remove_constant Man page
pre_remove_correlated Man page
pre_scale Man page
pre_split Man page
pre_transpose Man page
print.preprocessed_data Man page
pvalue Man page
pvalue.coxph Man page
pvalue.crr Man page
pvalue.cuminc Man page
pvalue.survdiff Man page
resample Man page
resample_bootstrap Man page
resample_crossvalidation Man page
resample_holdout Man page
reset_notification Man page
rmse Man page
roc_curve Man page
select Man page
select_.list Man page
select_.modeling_result Man page
subresample Man page
subtree Man page
trivial_error_rate Man page
tune Man page
validate_data Man page
vlines Man page
weighted_error_rate Man page


emil/man/index_fit.Rd emil/man/fit_randomForest.Rd emil/man/pre_impute.Rd emil/man/predict_lda.Rd emil/man/fit.Rd emil/man/nice_axis.Rd emil/man/pipe.Rd emil/man/predict_glmnet.Rd emil/man/tune.Rd emil/man/subtree.Rd emil/man/predict_coxph.Rd emil/man/predict.model.Rd emil/man/pvalue.survdiff.Rd emil/man/fit_pamr.Rd emil/man/error_fun.Rd emil/man/trivial_error_rate.Rd emil/man/learning_curve.Rd emil/man/weighted_error_rate.Rd emil/man/fit_lda.Rd emil/man/get_response.Rd emil/man/subresample.Rd emil/man/roc_curve.Rd emil/man/evaluate.Rd emil/man/get_tuning.Rd emil/man/predict_pamr.Rd emil/man/fit_qda.Rd emil/man/importance_randomForest.Rd emil/man/factor_to_logical.Rd emil/man/print.preprocessed_data.Rd emil/man/importance_pamr.Rd emil/man/nice_require.Rd emil/man/pre_pamr.Rd emil/man/is_constant.Rd emil/man/pvalue.crr.Rd emil/man/is_multi_procedure.Rd emil/man/get_prediction.Rd emil/man/validate_data.Rd emil/man/predict_cforest.Rd emil/man/plot.learning_curve.Rd emil/man/get_importance.Rd emil/man/na_index.Rd emil/man/dichotomize.Rd emil/man/resample.Rd emil/man/fit_glmnet.Rd emil/man/pre_impute_knn.Rd emil/man/fit_naive_bayes.Rd emil/man/pre_log_message.Rd emil/man/extension.Rd emil/man/is_blank.Rd emil/man/indent.Rd emil/man/mode.Rd emil/man/get_performance.Rd emil/man/neg_gmpa.Rd emil/man/image.resample.Rd emil/man/predict_caret.Rd emil/man/pre_impute_df.Rd emil/man/predict_randomForest.Rd emil/man/name_procedure.Rd emil/man/impute.Rd emil/man/vlines.Rd emil/man/fit_cforest.Rd emil/man/fit_rpart.Rd emil/man/plot.Surv.Rd emil/man/pvalue.coxph.Rd emil/man/pre_factor_to_logical.Rd emil/man/pre_process.Rd emil/man/fill.Rd emil/man/as.modeling_procedure.Rd emil/man/modeling_procedure.Rd emil/man/get_color.Rd emil/man/fit_coxph.Rd emil/man/fit_caret.Rd emil/man/predict_naive_bayes.Rd emil/man/notify_once.Rd emil/man/nice_box.Rd emil/man/importance_glmnet.Rd emil/man/predict_qda.Rd emil/man/pvalue.cuminc.Rd emil/man/fit_lm.Rd emil/man/predict_rpart.Rd emil/man/list_method.Rd emil/man/predict_lm.Rd emil/man/pvalue.Rd emil/man/log_message.Rd emil/man/predict_svm.Rd emil/man/select.Rd emil/man/fit_svm.Rd emil/man/emil.Rd

Questions? Problems? Suggestions? or email at ian@mutexlabs.com.

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.