interactive_text_explanations: Interactive explanations

In lime: Local Interpretable Model-Agnostic Explanations

Interactive explanations

Description

Display text explanation in an interactive way. You can :

Create an output to insert text explanation plot in Shiny application.

Render the text explanations in Shiny application.

Usage

interactive_text_explanations(
  explainer,
  window_title = "Text model explainer",
  title = "Local Interpretable Model-agnostic Explanations",
  place_holder = "Put here the text to explain",
  minimum_lentgh = 3,
  minimum_lentgh_error = "Text provided is too short to be explained (>= 3).",
  max_feature_to_select = 20
)

text_explanations_output(outputId, width = "100%", height = "400px")

render_text_explanations(expr, env = parent.frame(), quoted = FALSE)

Arguments

explainer	parameters
window_title, title, place_holder, minimum_lentgh_error	text to be displayed on the page
minimum_lentgh	don't update display if text is shorter than this parameter
max_feature_to_select	up limit to the number of words that can be selected
outputId	output variable to read from
width, height	Must be a valid CSS unit or a number, which will be coerced to a string and have "px" appended.
expr	An expression that generates an HTML widget
env	The environment in which to evaluate expr.
quoted	Is expr a quoted expression (with quote())? This is useful if you want to save an expression in a variable.

Details

• send a new sentence

• update the parameters of the explainer

Value

An output function that enables the use of the widget within Shiny applications.

A render function that enables the use of the widget within Shiny applications.

Examples

## Not run: 
library(text2vec)
library(xgboost)

data(train_sentences)
data(test_sentences)

get_matrix <- function(text) {
  it <- itoken(text, progressbar = FALSE)
  create_dtm(it, vectorizer = hash_vectorizer())
}

dtm_train = get_matrix(train_sentences$text)

xgb_model <- xgb.train(list(max_depth = 7, eta = 0.1, objective = "binary:logistic",
                 eval_metric = "error", nthread = 1),
                 xgb.DMatrix(dtm_train, label = train_sentences$class.text == "OWNX"),
                 nrounds = 50)

sentences <- head(test_sentences[test_sentences$class.text == "OWNX", "text"], 1)
explainer <- lime(train_sentences$text, xgb_model, get_matrix)

# The explainer can now be queried interactively:
interactive_text_explanations(explainer)

## End(Not run)

lime documentation built on Aug. 19, 2022, 9:07 a.m.