Machine Learning Pre-trained Models¶
All existing pre-trained Machine Learning Models are listed here.
Unlike the Machine Learning Models found on the Machine Learning Models page, these pre-trained models cannot be trained or re-trained directly from IFS Cloud.
These models are pre-trained on a large and diverse dataset that is specifically designed and curated to support the functionality of the individual models.
There are three main models currently provided by the Machine Learning Service:
Business Card Analyzer¶
The Business Card Analyzer, defined as BUSINESS_CARD on this page, is a Machine Learning Model that enables users to seamlessly extract business card information by uploading an image of the card.
This model is pre-trained on a diverse dataset of business card images. The system can automatically scan the information through Optical Character Recognition, which is then presented for verification.
Once confirmed, the contact can be saved in the appropriate fields, categorizing them as a lead, prospect, customer, or any other relevant category.
Expense Receipt Automation¶
The Expense Receipt Automation, defined as EXPENSE_RECEIPT on this page, is a Machine Learning Model that allows users to easily capture and store expense data by simply uploading an image of their receipts.
This model is pre-trained on a large dataset of expense receipts of various types, allowing the system to swiftly scan the receipt to extract essential information such as the date, expense type, total amount (including taxes), etc. and provides it for user verification.
Once confirmed, these details are automatically saved to the relevant expense sheets, ensuring easy access and organization for future use.
Customer Order Automation¶
The Customer Order Automation, defined as INVOICE on this page, is a Machine Learning Model that allows users to effortlessly extract information from images and Purchase Order documents, automatically retrieving crucial details related to customer orders.
This model is pre-trained on a large dataset that includes data such as purchase orders, invoices, etc. that helps the model arrive at the scanned result.
This scanned information is then provided for user verification, after which the users can choose to create orders for them in release status, print or email the order confirmations, and thereby store the order details for easy access in the customer orders section of IFS Cloud.