Scribd Logo
Upload

Welcome to Scribd!

  • Combine Lang Model 1

    Uploaded by

    Chaos Diver
    12 views1 page
    This document provides information about the combine_lang_model command line tool: It generates a starter traineddata file that can be used to train an LSTM-based neural network model for optical character recognition (OCR). The tool takes in a unicharset file and optional wordlists as input and outputs a traineddata file without needing other preprocessing steps. It supports specifying the language, script directory, input unicharset file, and other options like whether the language is written right-to-left or how the recoder should handle character encoding.

    Original Description:

    Original Title

    COMBINE_LANG_MODEL_1

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    This document provides information about the combine_lang_model command line tool: It generates a starter traineddata file that can be used to train an LSTM-based neural network model for optical character recognition (OCR). The tool takes in a unicharset file and optional wordlists as input and outputs a traineddata file without needing other preprocessing steps. It supports specifying the language, script directory, input unicharset file, and other options like whether the language is written right-to-left or how the recoder should handle character encoding.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as pdf or txt
    12 views1 page

    Combine Lang Model 1

    Uploaded by

    Chaos Diver
    This document provides information about the combine_lang_model command line tool: It generates a starter traineddata file that can be used to train an LSTM-based neural network model for optical character recognition (OCR). The tool takes in a unicharset file and optional wordlists as input and outputs a traineddata file without needing other preprocessing steps. It supports specifying the language, script directory, input unicharset file, and other options like whether the language is written right-to-left or how the recoder should handle character encoding.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as pdf or txt
    of 1

    14.04.

    2023, 15:39 COMBINE_LANG_MODEL(1)

    COMBINE_LANG_MODEL(1) Manual Page

    NAME
    combine_lang_model - generate starter traineddata

    SYNOPSIS
    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]

    DESCRIPTION
    combine_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).

    OPTIONS
    --lang lang
    The language to use. Tesseract uses 3-character ISO 639-2 language codes.
    (See LANGUAGES)
    --script_dir PATH
    Directory name for input script unicharsets. It should point to the location of
    langdata (github repo) directory. (type:string default:)
    --input_unicharset FILE
    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:)
    --lang_is_rtl BOOL
    True if language being processed is written right-to-left (eg Arabic/Hebrew).
    (type:bool default:false)
    --pass_through_recoder BOOL
    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)

    file:///E:/self_projects/programming/libraries/Tesseract-OCR/combine_lang_model.1.html 1/2

    You might also like