How to Install and Uninstall tesseract-ocr-traineddata-chi_tra_vert Package on openSUSE Leap
Last updated: November 26,2024
1. Install "tesseract-ocr-traineddata-chi_tra_vert" package
In this section, we are going to explain the necessary steps to install tesseract-ocr-traineddata-chi_tra_vert on openSUSE Leap
$
sudo zypper refresh
Copied
$
sudo zypper install
tesseract-ocr-traineddata-chi_tra_vert
Copied
2. Uninstall "tesseract-ocr-traineddata-chi_tra_vert" package
This is a short guide on how to uninstall tesseract-ocr-traineddata-chi_tra_vert on openSUSE Leap:
$
sudo zypper remove
tesseract-ocr-traineddata-chi_tra_vert
Copied
3. Information about the tesseract-ocr-traineddata-chi_tra_vert package on openSUSE Leap
Information for package tesseract-ocr-traineddata-chi_tra_vert:
---------------------------------------------------------------
Repository : Main Repository
Name : tesseract-ocr-traineddata-chi_tra_vert
Version : 4.1.0+git.20191030.6572757-bp155.1.6
Arch : noarch
Vendor : openSUSE
Installed Size : 1.7 MiB
Installed : No
Status : not installed
Source package : tesseract-ocr-traineddata-4.1.0+git.20191030.6572757-bp155.1.6.src
Upstream URL : https://github.com/tesseract-ocr
Summary : Chinese (Traditional, Vertical) language data for tesseract-ocr-traineddata
Description :
This package contains the fast integer version of the Chinese (Traditional, Vertical) language
trained models for the Tesseract Open Source OCR Engine.
---------------------------------------------------------------
Repository : Main Repository
Name : tesseract-ocr-traineddata-chi_tra_vert
Version : 4.1.0+git.20191030.6572757-bp155.1.6
Arch : noarch
Vendor : openSUSE
Installed Size : 1.7 MiB
Installed : No
Status : not installed
Source package : tesseract-ocr-traineddata-4.1.0+git.20191030.6572757-bp155.1.6.src
Upstream URL : https://github.com/tesseract-ocr
Summary : Chinese (Traditional, Vertical) language data for tesseract-ocr-traineddata
Description :
This package contains the fast integer version of the Chinese (Traditional, Vertical) language
trained models for the Tesseract Open Source OCR Engine.