Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 2 Current »

Lazarus Riky


Author:  Jerry Chae

Overview

Lazarus Riky is a pre-trained Lazarus AI document processing solutions. It is similar to RikAI for the set of functions but Riky supports multi-language. It takes any document in 70+ languages and understands the context of it. The input file can be PDF, jpg, and png. Also, you can input your questions about the document – yes, Riky is also a natural language base conversational AI with computer vision and context analysis capabilities. The outputs is in general a text file that contains answers to the questions. As an advanced menu, the API also outputs additional information such as (invoice related) key-value pairs along with confidence level, text’s position coordinates, and size. You can also use a text file to feed a long list of questions.

To use this plugin, you must obtain API credentials from this link

https://api.lazarusforms.com/signup

For more information, please visit Lazarus Forms API Documentation page.

https://api.lazarusforms.com/docs?python#invoices


Need help?

Technical contact to tech@argos-labs.com


May you search all operations,




Input (Requirement)

  • orgId
  • authKey
  • Image File (PDF, jpeg, png, and Tiff

Input (Optional for additional infomation)

  • Questions in File (text file of questions with \n as a separator)
  • Encoding of choice such as UTF-8
  • JSON
  • YAML


Special Tags

  • For processing tabular data (tables and forms) there are two tags.
  • [FULL TABLE] and [TABLE] has to be used in all uppercase letters
  • When you are working with tabular data there you can add [FULL TABLE] or [TABLE] to the question, in order to toggle the model into table mode. In its current state, the model functions best with Korean and Japanese languages. English is non-optimal in its current state, we do not recommend demonstrating in English.


here is an example:

--form 'question="[TABLE]주민등록번호는?"'


Return Value

Answers to the questions in text format in String, CSV, or File

Return Codes

  • 0 Success
  • 1 Invalid Org ID or Auth Key
  • 2 Invalid input image file format
  • 99 misc. errors


Parameter Setting samples



  • No labels