📄 Extract Sentiment¶

Note

ID: templates.targeted_sentiment_extraction.extract_sentiment | Type: SpanLabelingTemplate

{
    "__type__": "span_labeling_template",
    "input_format": "From the following {text_type}, extract the objects for which the sentiment expressed is positive, and the objects for which the expressed sentiment is negative, and the objects for which the expressed sentiment is neutral.\n{text_type}: {text}\n",
    "postprocessors": [
        "processors.to_span_label_pairs"
    ]
}

References: processors.to_span_label_pairs