Start using Extract Beta
Requirements
Write to [email protected] to request the activation of Extract API for your developer account.
Python client
expertai-extract is a Python package corresponding to the extract-pyton
GitHub project.
Follow the instructions on the GitHub project home page to install the package and start using Extract Beta.
For the interpretation of the output, read the dedicated article in the reference section of this manual.
Experiment with Postman
If, instead of using the Python client which saves you from knowing the low-level details, you want to experiment with HTTP requests and responses, you can use Postman and download the sample collection.
Steps:
- Run Postman.
- Import the sample collection.
-
Set collection variables:
- Double-click the collection name on the left panel.
- Select the Variables tab.
- Set the current value for
username
andpassword
to you expert.ai developer account credentials. - Select Save on the toolbar.
-
Open the
token
request and select Send. This will automatically provide necessary authorizations. - Open one of the
layout-document-async sample...
requests and select Send. This will start the layout recognition task for a sample document. - Open the
status
request and select Send until you get the results (the value of thecurrent
key must reach100
). Interpret the results with the help of the dedicated article.
You can repeat from step 5 with another layout-document-async sample...
request to experiment with other sample documents.
If you want to experiment with your own documents:
- Duplicate one of the
layout-document-async sample ...
requests. - Open the duplicate request and select Body.
- Change the value of the
name
key to the name of your PDF file. - Change the value of the
data
key to the Base64 representation of your PDF file (you can easily find free utilities on the Web to create the Base64 representation of a file). - Save the request.
- Select Send.
- Open the
status
request and select Send until you get the results (the value of thecurrent
key must reach100
). Interpret the results with the help of the dedicated article.
Create a custom client
The Extract Beta API can be consumed with a variety of programming languages and ready-made utilities. To create client code tailored to your need with the language you prefer, read the next article.