Invoice with metadata

Extracting additional metadata from invoices with examples in Python and Node.

Overview

This guide covers how to extract metadata from multiple supplier invoices with examples in Python and Node. You will extract the following fields:

Name of the supplier
Name of the receiver
Invoice number
Purchase order number
Issue date
Pay due date
Total amount

This guide shows you how to

Get your API Key

The Authorization header for your API key is: Token YOUR-API-KEY (Login if you do not see one). You can also obtain the API key by visiting the Settings page.

1. Create a new document type

Before you start extracting data, you need to define a document type. Navigate to the Dashboard page and click on the New document type button in the top right corner of the table. Next, select the Metadata invoice card. The wizard will already pre-fill all the needed extraction fields along with the document type configuration. Click on the Create document type. This will create a new document type with name metadata-invoice with the following fields:

supplier_name
invoice_number
purchase_order_number
receiver_name
issue_date
pay_due_date
total_amount

2. Add suppliers

Typless is a tool for automation. That's why you need to fill the dataset and train it first. To automate a new supplier, you need to first add its invoices to the data set. Download invoices - one from Awesome Company and another from Good Services.

To add a document to the dataset, use the add-document endpoint or use the training room, where you can easily upload a file and fill out the necessary information.

The dataset is created by uploading an original file with the correct value for each field defined inside the document type:

1 Open file as base64 string (Lines 6-7)

Make sure you are pointing to the correct path in your file system.

2 Specify payload (Lines 9-43)

3 Specify headers (Lines 48-52)

4 Send POST request (Lines 54-56)

import requests
import base64

file_name = 'amazing_company_1.pdf'

with open(file_name, 'rb') as file:
    base64_data = base64.b64encode(file.read()).decode('utf-8')

payload = {
    "file": base64_data,
    "file_name": file_name,
    "document_type_name": "metadata-invoice",
    "learning_fields": [
        {
            "name": "supplier_name",
            "value": "Amazing Company"
        },
        {
            "name": "receiver_name",
            "value": "Amazing Client"
        },
        {
            "name": "invoice_number",
            "value": "333"
        },
        {
            "name": "purchase_order_number",
            "value": "234778"
        },
        {
            "name": "pay_due_date",
            "value": "2021-03-31"
        },
        {
            "name": "issue_date",
            "value": "2021-02-01"
        },
        {
            "name": "total_amount",
            "value": "15.0000"
        }
    ]
}


url = "https://developers.typless.com/api/add-document"

headers = {
    "Accept": "application/json",
    "Content-Type": "application/json",
    "Authorization": "<<apiKey>>"
}

response = requests.request("POST", url, json=payload, headers=headers)

print(response.json())

const fetch = require('node-fetch');
const fs = require('fs');

const fileName = 'amazing_company_1.pdf';
const base64File = fs.readFileSync(fileName, {encoding: 'base64'});

let url = 'https://developers.typless.com/api/add-document';

const payload = {
  file: base64File,
  file_name: fileName,
  document_type_name: "metadata-invoice",
  learning_fields: [
      {
          name: "supplier_name",
          value: "Amazing Company"
      },
      {
          name: "receiver_name",
          value: "Amazing Client"
      },
      {
          name: "invoice_number",
          value: "333"
      },
      {
          name: "purchase_order_number",
          value: "234778"
      },
      {
          name: "pay_due_date",
          value: "2021-03-31"
      },
      {
          name: "issue_date",
          value: "2021-02-01"
      },
      {
          name: "total_amount",
          value: "15.0000"
      }
  ]
};

const headers = {
  'Accept': 'application/json',
  'Content-Type': 'application/json',
  'Authorization': '<<apikey>>'
}

let options = {
  method: 'POST',
  headers: headers,
  body: JSON.stringify(payload)
};

fetch(url, options)
  .then(res => res.json())
  .then(json => console.log(json))
  .catch(err => console.error('error:' + err));

Response:

 {
    "details":["0cb9c9227a1f67dcd54ce4019aed6867a8b32a5b"],
    "message":"Document added successfully."
}

As you can see, to achieve high accuracy, Typless only needs the values that are in the document. Nevertheless, there are some rules to keep in mind when providing values. Applying these to the Amazing Company example, we changed three fields:

total_amount value was converted with number type rules from 15,00 to 15.0000
issue_date value was converted with date type rules from the word Feb 1, 2021, to 2021-02-01
pay_due_date value was converted with date type rules from the word Mar 31, 2021, to 2021-03-31

The same rules were also applied to the Good Services example.

You will have two suppliers added to your document type after you run both code examples.

3. Execute training

👍 Training is executed automatically every day at 10 PM CET

For all of your suppliers with new documents in the dataset of all your document types. Free of charge

To immediately see results, you can trigger the training process on the Dashboard page. Look for the metadata-invoice document type in the list, and click on . The training for the example should finish in a matter of seconds.

Need more information? Read more about training.

4. Extract data from documents

After the training is finished, you can start precisely extracting data from documents from trained suppliers. Here you have two new invoices from the trained suppliers:

Download them and extract the data using the code:

1 Open file as base64 string (Lines 5-7)

Make sure you are pointing to the correct path in your file system.

2 Specify payload (Lines 9-12)

3 Specify headers (Lines 17-20)

4 Send POST request (Line 23)

import requests
import requests
import base64

file_name = 'amazing_company_2.pdf'
with open(file_name, 'rb') as file:
    base64_data = base64.b64encode(file.read()).decode('utf-8')

payload = {
    "file": base64_data,
    "file_name": file_name,
    "document_type_name": "metadata-invoice"
}

url = "https://developers.typless.com/api/extract-data"

headers = {
    "Accept": "application/json",
    "Content-Type": "application/json",
    "Authorization": "<<apikey>>"
}

response = requests.request("POST", url, json=payload, headers=headers)

for field in response.json()['extracted_fields']:
    print(f'{field["name"]}: {field["values"][0]["value"]}')

const fetch = require('node-fetch');
const fs = require('fs');

const fileName = 'amazing_company_2.pdf';
const base64File = fs.readFileSync(fileName, {encoding: 'base64'});

const url = 'https://developers.typless.com/api/extract-data';

const payload = {
  file: base64File,
  file_name: fileName,
  document_type_name: "metadata-invoice"
}

const headers = {
  'Accept': 'application/json',
  'Content-Type': 'application/json',
  'Authorization': '<<apikey>>'
}

let options = {
  method: 'POST',
  headers: headers,
  body: JSON.stringify(payload)
};

fetch(url, options)
  .then(res => res.json())
  .then(json => {
    json.extracted_fields.forEach(field => console.log(`${field.name}: ${field.values[0].value}`))
  })
  .catch(err => console.error('error:' + err));

Response:

// Example extraction response - the provided recipe will not produce equal results
{
    "file_name": "invoice_2.pdf",
    "object_id": "1cb25cc8-c9fa-4149-9a83-b4ed6a2173b9",
    "extracted_fields": [
        {
            "name": "supplier_name",
            "values": [
                {
                    "x": -1,
                    "y": -1,
                    "width": -1,
                    "height": -1,
                    "value": "ScaleGrid",
                    "confidence_score": "0.968",
                    "page_number": -1
                }
            ],
            "data_type": "AUTHOR"
        },
        {
            "name": "invoice_number",
            "values": [
                {
                    "x": 1989,
                    "y": 545,
                    "width": 323,
                    "height": 54,
                    "value": "20190500005890",
                    "confidence_score": "0.250",
                    "page_number": 0
                },
                {
                    "x": 167,
                    "y": 574,
                    "width": 391,
                    "height": 54,
                    "value": "GB123456789",
                    "confidence_score": "0.250",
                    "page_number": 0
                }
            ],
            "data_type": "STRING"
        },
        {
            "name": "issue_date",
            "values": [
                {
                    "x": 2072,
                    "y": 628,
                    "width": 240,
                    "height": 54,
                    "value": "2019-06-05",
                    "confidence_score": "0.358",
                    "page_number": 0
                }
            ],
            "data_type": "DATE"
        },
      	{
            "name": "pay_due_date",
            "values": [
                {
                    "x": 2072,
                    "y": 628,
                    "width": 240,
                    "height": 54,
                    "value": "2019-06-05",
                    "confidence_score": "0.358",
                    "page_number": 0
                }
            ],
            "data_type": "DATE"
        },
        {
            "name": "total_amount",
            "values": [
                {
                    "x": 2146,
                    "y": 1196,
                    "width": 126,
                    "height": 54,
                    "value": "47.5300",
                    "confidence_score": "0.990",
                    "page_number": 0
                }
            ],
            "data_type": "NUMBER"
        }
    ],
    "line_items": []
}
    "customer": null
}

Need a more in-depth explanation of the response? You can read about it here.

5. Continuously improve models

Typless embraces the fact that the world is changing all the time. That's why you can improve models on the fly by providing correct data after extraction. Let's say your company has a new partner, Best Supplier. You don't need to start over with building the dataset. You can simply extract and send the correct data after they are verified by your users. You can learn more about providing feedback on the building dataset page.

📘 Closed workflow loop - improve models live!

Use every action from your users to adapt and improve Typless models without any extra costs.

To send feedback, use the add-document-feedback with the object_id.

Running Typless live

The only thing that you need to do to automate your manual data entry is to integrate those simple API calls into your system.

Have any questions, or do you need some help? Contact us via email support@typless.com.

PreviousOverview NextInvoice with line items

Last updated 1 month ago