Image Classifier

Quick Start

To get started:

• Choose a trained model from the Model to use dropdown
• Send an image path via msg.payload
• Receive the classification result in msg.payload

Configuration

Model to use (required)

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

Common Input Format (All Algorithms)

msg.payload (string)

Relative path of the image file on shared storage.

Example: "images/photo.jpg" or "documents/image.png"

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

Output by Algorithm

Algorithm 1 / Algorithm 2

msg.payload is an object with prediction and confidence:

Example: {"prediction": "invoice", "confidence": "0.98"}

Algorithm 3

msg.payload is an object with prediction only (no confidence):

Example: {"prediction": "invoice"}

Optional: model-specific dimensions

Some models require you to provide the expected input width and height.

Example

Input (msg.payload)

"images/photo.jpg"

Output (msg.payload)

{
  "prediction": "product_photo",
  "confidence": "0.89"
}

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 must be a valid image path on shared storage.
• Unsupported image type: Only .png, .jpg, and .jpeg are supported.
• Dimensions don’t match model expectations: If your selected model requires width/height, set them to match the model input shape.

Best Practices

• Use clear, well-lit images with appropriate resolution for better accuracy
• Ensure your training data covers all expected image categories comprehensively
• Regularly retrain and update models as your classification categories evolve
• Always validate classification results in production applications, especially for critical decisions