README.md

In andrew-mcinerney/statnn: Interpreting Neural Networks From a Statistical Perspective

interpretnn

Installation

You can install the current version of interpretnn from CRAN with:

install.packages("interpretnn")

You can install the development version of interpretnn from GitHub with:

# install.packages("devtools")
devtools::install_github("andrew-mcinerney/interpretnn")

interpretnn()

The primary function in this package is interpretnn(). It creates a more statistically-based object of an existing neural network object. Currently supports neural networks from nnet, neuralnet, keras, ANN, and torch.

library(interpretnn)
intnn <- interpretnn(object)

A useful summary table can be generated using

summary(intnn)

and covariate-effect plots can be created using

plot(intnn, conf_int = TRUE)

More information about these functions and their arguments can be found in the documentation.

Notice

This package is currently in development. If you experience any issues, please get in touch.

andrew-mcinerney/statnn documentation built on June 30, 2024, 4:09 p.m.