Document Classifier

Quick Start

To get started:

• Choose an option from the Choose Algorithm dropdown
• Choose a trained model from the Model to use dropdown
• Send document image path via msg.payload.image_path
• Optionally provide OCR data via msg.payload.words_and_bboxes for better accuracy
• Receive the document type in msg.payload

Configuration

Model to use (required)

Select a pre-trained model from the dropdown menu. Models must be trained beforehand using the document classifier trainer block.

Common Input Format (All Algorithms)

msg.payload.image_path (string)

Relative path of the document image file on shared storage.

Example: "documents/invoice.png" or "scans/receipt.jpg"

Supported formats: .png, .jpg, .jpeg (case insensitive)

msg.payload.words_and_bboxes (array, optional)

Optional OCR data containing words and their bounding boxes. Providing this can improve classification accuracy.

Format: [[[x1, y1, x2, y2], "word"], ...]

Example: [[[29, 23, 150, 45], "Invoice"], [[29, 50, 200, 70], "Date"]]

Note: Coordinates are [top-left-x, top-left-y, bottom-right-x, bottom-right-y]

Common Output Format (All Algorithms)

msg.payload (object)

msg.payload contains an output field with the predicted document type.

Example: {"output": "invoice"}

Example

Input (msg.payload)

{
  "image_path": "documents/invoice.png",
  "words_and_bboxes": [
    [[29, 23, 150, 45], "Invoice"],
    [[29, 50, 200, 70], "Date"]
  ]
}

Output (msg.payload)

{
  "output": "invoice"
}

Errors

When the block fails, it raises an error. Use a Catch block in your flow to handle failures and inspect the error payload.

Common mistakes

• Missing or wrong image path: msg.payload.image_path is required and must point to a file on shared storage.
• Unsupported image type: Only .png, .jpg, and .jpeg are supported.
• Wrong OCR structure: msg.payload.words_and_bboxes must follow [[[x1, y1, x2, y2], "word"], ...].

Best Practices

• Use clear, well-scanned document images for better classification accuracy
• Provide OCR data (words_and_bboxes) whenever available to improve classification results
• Ensure your training data includes diverse examples of each document type
• Regularly retrain models as new document types or variations are encountered
• Always validate classification results in production applications, especially for critical workflows