Using Automation 360 IQ Bot Cloud
This topic guides you through the process of creating, training, extracting data from, and validating a new learning instance.
Complete the steps to install IQ Bot.
In Automation 360 IQ Bot, click the Learning
Instance tab to create a new learning instance and upload sample
documents to train.
Create a learning instanceNote: As Tegaki API OCR requires a separate On-Premises set up that is not supported in Automation 360 IQ Bot Cloud, all other OCR engines except the Tegaki API OCR are available.
- You will always have the latest version of the OCR engines supported by Automation 360 IQ Bot Cloud, but cannot select a specific OCR version.
- You cannot configure OCR settings from the Control Room. Starting with Automation 360 IQ Bot.19 version, we have provided an internal API that can be used to configure OCR settings in the database.
Train the uploaded sample documents in the IQ Bot
Review the supported Pandas libraries:
During the training, you can add and use custom logic in the Designer to improve data extraction in production.
After training is complete, the learning instance is ready for production.
In Automation Anywhere Control Room, upload documents for processing the images
and extracting the required fields into a CSV file using the IQ Bot
Upload document action on the Bot Runner
The IQ Bot server analyzes the uploaded documents using an AI model.
Files are uploaded from the local folder on the workstation to the IQ Bot server through a secure HTTPS session. Uploaded production documents are deleted after successful processing.
These uploaded documents are stored in a database. The data is encrypted and cannot be accessed by anyone.
- Download the extracted data to a CSV file using the Download all documents action. In addition to the successfully extracted values, this action also downloads any unclassified, untrained, and invalid documents.
- Unclassified and invalid documents are moved to a separate folder for manual validation.