shahryary / deepoperators

Deeply Learned Boolean Operators

DeepOperators

The goal of DeepOperators is to provide pre-trained deeply learned boolean operators.

The %&% and %|% operators turn plain old business logic into “deep learning”, “machine learning”, and “AI” problems. Because business folk sometimes dictate how problems are solved, DeepOperators enables the programmer Data Scientist to use “deep learning” when it is required but it would otherwise be more practical to use built-in logical operators.

Installation

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

# install.packages("remotes")
remotes::install_github("ellisvalentiner/DeepOperators")

Example

This is a basic example of the DeepOperator functions:

library(DeepOperators)

# using the deep AND operator, %&%
TRUE %&% FALSE
#> [1] FALSE
TRUE %&% TRUE
#> [1] TRUE

# using the deep OR operator, %|%
TRUE %|% FALSE
#> [1] TRUE
FALSE %|% FALSE
#> [1] FALSE

Additionally DeepOperator provides function to automatically re-train the deep operators.

train_deep_or()
train_deep_and()

Provenance

This package was inspired by Fizz Buzz in Tensorflow.