Invoices Japan - ML package

Overview

An invoice, bill or tab is a commercial document issued by a seller to a buyer in a sale transaction and indicates the products, quantities, and agreed-upon prices for products or services the seller had provided the buyer.

The Invoices models are trained to extract key data points from various types of invoices. The schema consists of 48 fields, of which 10 items, and a currency classifier extracting information about customers, vendors, payments and products/ services, payment details including bank info. Customers can adjust the schema of fields to their own requirements.

Particularities

The model automatically detects data such as:

• Vendor Name, Vendor Address, Billing Name, Billing Address
• Invoice Number, Purchase Order Number, Payment Terms, Due Date
• Net Amount, Net Amount Reduced, Net Amount Non-reduced
• Tax Amount, Tax Reduced, Tax Non-reduced, Witholding Tax Amount, Discount, Total Amount
• Item Tax Rate, Item Registration Tax, Item Fee
• Shipping Address, Shipping Date, Shipping Method
• VAT Number, VAT Rate
• Bank Name, Bank Account Number, Bank Branch Name, Bank Branch Number, Bank Account Type, Bank Account Name

Note: For the complete list of extracted fields, check the Out-of-the-box Models Details file.

Samples

Here are some samples of an Invoice form:

Related out-of-the-box pre-trained models:

• Invoices
• InvoicesChina
• InvoicesIndia
• Invoices Shipping

Processed document types:

This out-of-the-box pre-trained model can be used with the following document types:

• Invoices
• Credit Notes