People

Identify references to specific persons found in a document.

Current Version: 2

Output

This function will return a list of dictionaries of three key-value pairs. Each dictionary represents a substring from the input document that the model has predicted is a person with confidence greater than your threshold. Each dictionary has three keys: text, confidence, and position. The text key maps to a substring from the document that is predicted to be a person. The confidence key maps to a float between 0 and 1, representing how confident the model is that the text is a person. The position key maps to a two element list with the start character position and end character position of text in the input document.

# single output
[
    {
        'text': 'Mike Brown',
        'confidence':  0.913224,
        'position':  [26, 36]
    }
]

# batch output
[
    [
        {
            'text': 'Mike Brown',
            'confidence':  0.913224,
            'position':  [26, 36]
        }
    ],
    [
        {
            'text': 'Mike Brown',
            'confidence':  0.913224,
            'position':  [26, 36]
        }
    ]
]

Example