1COMBINE_LANG_MODEL(1) 2===================== 3:doctype: manpage 4 5NAME 6---- 7combine_lang_model - generate starter traineddata 8 9SYNOPSIS 10-------- 11*combine_lang_model* --input_unicharset 'filename' --script_dir 'dirname' --output_dir 'rootdir' --lang 'lang' [--lang_is_rtl] [pass_through_recoder] [--words file --puncs file --numbers file] 12 13DESCRIPTION 14----------- 15combine_lang_model(1) generates a starter traineddata file that can be used to train an LSTM-based neural network model. It takes as input a unicharset and an optional set of wordlists. It eliminates the need to run set_unicharset_properties(1), wordlist2dawg(1), some non-existent binary to generate the recoder (unicode compressor), and finally combine_tessdata(1). 16 17OPTIONS 18------- 19'--lang lang':: 20 The language to use. 21 Tesseract uses 3-character ISO 639-2 language codes. (See LANGUAGES) 22 23'--script_dir PATH':: 24 Directory name for input script unicharsets. It should point to the location of langdata (github repo) directory. (type:string default:) 25 26'--input_unicharset FILE':: 27 Unicharset to complete and use in encoding. It can be a hand-created file with incomplete fields. Its basic and script properties will be set before it is used. (type:string default:) 28 29'--lang_is_rtl BOOL':: 30 True if language being processed is written right-to-left (eg Arabic/Hebrew). (type:bool default:false) 31 32'--pass_through_recoder BOOL':: 33 If true, the recoder is a simple pass-through of the unicharset. Otherwise, potentially a compression of it by encoding Hangul in Jamos, decomposing multi-unicode symbols into sequences of unicodes, and encoding Han using the data in the radical_table_data, which must be the content of the file: langdata/radical-stroke.txt. (type:bool default:false) 34 35'--version_str STRING':: 36 An arbitrary version label to add to traineddata file (type:string default:) 37 38'--words FILE':: 39 (Optional) File listing words to use for the system dictionary (type:string default:) 40 41'--numbers FILE':: 42 (Optional) File listing number patterns (type:string default:) 43 44'--puncs FILE':: 45 (Optional) File listing punctuation patterns. The words/puncs/numbers lists may be all empty. If any are non-empty then puncs must be non-empty. (type:string default:) 46 47'--output_dir PATH':: 48 Root directory for output files. Output files will be written to <output_dir>/<lang>/<lang>.* (type:string default:) 49 50HISTORY 51------- 52combine_lang_model(1) was first made available for tesseract4.00.00alpha. 53 54RESOURCES 55--------- 56Main web site: <https://github.com/tesseract-ocr> + 57Information on training tesseract LSTM: <https://tesseract-ocr.github.io/tessdoc/TrainingTesseract-4.00.html> 58 59SEE ALSO 60-------- 61tesseract(1) 62 63COPYING 64------- 65Copyright \(C) 2012 Google, Inc. 66Licensed under the Apache License, Version 2.0 67 68AUTHOR 69------ 70The Tesseract OCR engine was written by Ray Smith and his research groups 71at Hewlett Packard (1985-1995) and Google (2006-present). 72