Usage
Installation
The DocsQuality prediction engine may be installed on the following versions:
standalone (offline),
web (online).
The sections below describe the detailed process with integration examples.
Offline integration
Important
Prepare license file available in the user account or from the internal administrator.
Linux
Install required dependencies
Ensure your system has the necessary dependencies.
sudo apt-get update
Debian/Ubuntu (apt):
sudo apt-get install binutils libgl1-mesa-glx libglib2.0-0 libsm6 libice6
CentOS Stream 9 (yum):
sudo yum install binutils mesa-libGL glib2
Fedora (dnf):
sudo dnf install binutils libGL
Install Tesseract OCR
Note
Tesseract is used to predict the ocr index and document category.
Install Tesseract OCR using the following command (Ubuntu):
apt-get install tesseract-ocr libtesseract-dev
Set environment variable TESSDATA_PREFIX to include the path to the tessdata folder.
export TESSDATA_PREFIX=path/to/tesseract-ocr/version/tessdata
Example:
export TESSDATA_PREFIX=/usr/share/tesseract-ocr/5/tessdata
For more information, please visit https://tesseract-ocr.github.io/tessdoc/Installation.html.
Download and Unpack the Package
Download the desired package from the Binaries section in your account.
Unpack the downloaded package using the following command:
tar -xvf package_name.tar.gz
Place the License File
Move the prepared license file to the main directory where the application is extracted.
Run the Application
Navigate to the directory where the application is located and execute:
cd path/to/application ./server
Windows
Note
Windows packages available upon request. If necessary, please contact us at support@docsquality.com
Download and Unpack the Package
Download the desired package from the Binaries section in your account.
Unpack the downloaded package.
Place the License File
Move the prepared license file to the main directory where the application is extracted.
Run the Application
Click on the executable file to run the application.
Online integration
Important
Prepare client credentials available in the user account or from the internal administrator.
Steps to integrate:
1. Firstly, using the client’s credentials, the server must obtain an access token by communicating with the identity server.
Note
The token also contains information about its expiration.
Using the received access token, send the document file for prediction.
For integration examples across various programming languages, please visit our Github.
Detailed information about making requests to the API endpoints can be found here.