compare_all: Compare All

In soodoku/recognize: Assess Quality of OCR

Description

Takes path to directory with OCRd files and path to directory with human transcribed files and calculates error rate. Calls compare_txt

Usage

compare_all(ocr_dir = NULL, gold_dir = NULL, remove_extra_space = TRUE,
  normalize = TRUE)

Arguments

ocr_dir	path to directory containing files with OCRd text; required
gold_dir	path to directory containing files with human transcribed text; required
remove_extra_space	a dummy indicating whether or not extra spaces should be removed from the OCR file; default is TRUE
normalize	normalize string distance measures — otherwise longer the document, greater the distance

Value

error rate

Examples

## Not run: 
compare_all(ocr_dir="path_to_ocr_dir", gold_dir="path_to_gold_dir", 
	           remove_extra_space=TRUE)

## End(Not run)

soodoku/recognize documentation built on May 30, 2019, 6:30 a.m.