Title: Word Factor Vectors
Version: 0.0.1
Description: A user-friendly factor-like interface for converting strings of text into numeric vectors and rectangular data structures.
Encoding: UTF-8
LazyData: true
Imports: xgboost, tokenizers, text2vec, R6, utils, tibble, ggplot2, stats, Matrix
URL: https://github.com/mkearney/wactor
BugReports: https://github.com/mkearney/wactor/issues
RoxygenNote: 7.0.2
License: MIT + file LICENSE
Suggests: testthat (≥ 2.1.0), covr
NeedsCompilation: no
Packaged: 2019-12-13 05:40:20 UTC; kmw
Author: Michael W. Kearney ORCID iD [aut, cre], Lingshu Hu ORCID iD [ctb]
Maintainer: Michael W. Kearney <kearneymw@missouri.edu>
Repository: CRAN
Date/Publication: 2019-12-18 15:30:02 UTC

A wactor object

Description

A factor-like class for word vectors

Methods

Public methods

Method new()

Usage
Wactr$new(
  text = character(),
  tokenizer = NULL,
  max_words = 1000,
  doc_prop_max = 1,
  doc_prop_min = 0
)
Arguments
max_words

Maximum number of words in vocabulary

doc_prop_max

Maximum proportion of docs for terms in dinctionary

doc_prop_min

Minimum proportion of docs for terms in dictionary.

Method clone()

The objects of this class are cloneable with this method.

Usage
Wactr$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

As wactor

Description

Convert data into object of type 'wactor'

Usage

as_wactor(.x, ...)

Arguments

.x

Input text vector

...

Other args passed to Wactr$new(...)

Value

An object of type wactor

Document term frequency

Description

Converts character vector into document term matrix (dtm)

Usage

dtm(object, .x = NULL)

Arguments

object

Input object containing dictionary (column), e.g., wactor

.x

Text from which the document term matrix will be created

Value

A c-style matrix

Examples


## create wactor
w <- wactor(letters)

## use wactor to create dtm of same vector
dtm(w, letters)

## using the initial data is the default; so you don't actually have to
## respecify it
dtm(w)

## use wactor to create dtm on new vector
dtm(w, c("a", "e", "i", "o", "u"))

## apply directly to character vector
dtm(letters)

Split into test and train data sets

Description

Randomly partition input into a list of train and test data sets

Usage

split_test_train(.data, .p = 0.8, ...)

Arguments

.data

Input data. If atomic (numeric, integer, character, etc.), the input is first converted to a data frame with a column name of "x."

.p

Proportion of data that should be used for the train data set output. The default value is 0.80, meaning the train output will include roughly 80 pct. of the input cases while the test output will include roughly 20 oct..

...

Optional. The response (outcome) variable. Uses tidy evaluation (quotes are not necessary). This is only relevant if the identified variable is categorical–i.e., character, factor, logical–in which case it is used to ensure a uniform distribution for the train output data set. If a value is supplied, uniformity in response level observations is prioritized over the .p (train proportion) value.

Value

A list with train and test tibbles (data.frames)

Examples


## example data frame
d <- data.frame(
  x = rnorm(100),
  y = rnorm(100),
  z = c(rep("a", 80), rep("b", 20))
)

## split using defaults
split_test_train(d)

## split 0.60/0.40
split_test_train(d, 0.60)

## split with equal response level obs
split_test_train(d, 0.80, label = z)

## apply to atomic data
split_test_train(letters)

Term frequency inverse document frequency

Description

Converts character vector into a term frequency inverse document frequency (TFIDF) matrix

Usage

tfidf(object, .x = NULL)

Arguments

object

Input object containing dictionary (column), e.g., wactor

.x

Text from which the tfidf matrix will be created

Value

A c-style matrix

Examples


## create wactor
w <- wactor(letters)

## use wactor to create tfidf of same vector
tfidf(w, letters)

## using the initial data is the default; so you don't actually have to
## respecify it
tfidf(w)

## use wactor to create tfidf on new vector
tfidf(w, c("a", "e", "i", "o", "u"))

## apply directly to character vector
tfidf(letters)

Create wactor

Description

Create an object of type 'wactor'

Usage

wactor(.x, ...)

Arguments

.x

Input text vector

...

Other args passed to Wactr$new(...)

Value

An object of type wactor

Examples


## create
w <- wactor(c("a", "a", "a", "b", "b", "c"))

## summarize
summary(w)

## plot
plot(w)

## predict
predict(w)

## use on NEW data
dtm(w, letters[1:5])

## dtm() is the same as predict()
predict(w, letters[1:5])

## works if you specify 'newdata' too
predict(w, newdata = letters[1:5])

xgb matrix

Description

Simple wrapper for creating a xgboost matrix

Usage

xgb_mat(x, ..., y = NULL, split = NULL)

Arguments

x

Input data

...

Other data to cbind

y

Label vector

split

Optional number between 0-1 indicating the desired split between train and test

Value

A xgb.Dmatrix

Examples


xgb_mat(data.frame(x = rnorm(20), y = rnorm(20)))