b27a4754c8e6dd44821350037c036ff4eb061cb4,nala/learning/postprocessing.py,,predict_with_regex_patterns,#Any#,7

Before Change

    filter_short = re.compile("^[A-Z][0-9][A-Z]")

    for part_id, part in dataset.partids_with_parts():
        for regex in regex_patterns:
            for match in regex.finditer(part.text):
                offset = (match.start(), match.end(), part_id, "e_2")

After Change

    regex_patterns = construct_regex_patterns_from_predictions(dataset)

    existing_predictions = []
    for doc_id, doc in dataset.documents.items():
       for part_id, part in doc.parts.items():
           for ann in part.predicted_annotations:
                existing_predictions.append((ann.offset, ann.offset + len(ann.text), part_id, ann.class_id, doc_id))

    filter_short = re.compile("^[A-Z][0-9][A-Z]")

    for doc_id, doc in dataset.documents.items():
        for part_id, part in doc.parts.items():

Instances