Prodigy recipe to categorize differences in text

copyedits.jsonl

{"removed":"These results suggested that the distribution of bacterial communities was driven more by sample types than the separate caves from which samples were collected.","added":"These results suggest that the distribution of bacterial communities is driven more by sample types than the separate caves from which samples were collected.","meta":{"score":1}}

diff_text.py

import prodigy
from prodigy.components.loaders import JSONL

# run with 
# python -m prodigy diff-textcat copyedits copyedits.jsonl  -F .\diff_text.py

def add_label_to_stream(stream, label):
    for eg in stream:
        # The 'label' you get from the command line is a list
        # so let's just assume it's always one and take the first
        eg["label"] = label[0]
        yield eg

@prodigy.recipe(
    "diff-textcat",
    dataset=("The dataset to use", "positional", None, str),
    source=("The source data as a JSONL file", "positional", None, str),
    )
def copyedit(dataset, source):
    stream = JSONL(source)
    blocks = [
        {"view_id": "diff"},  
        {"view_id": "classification"},
        ]
    stream = stream
    return {
        "stream": stream,
        "dataset": dataset,
        "view_id": "blocks",
        "config": {
            "blocks": blocks,
        }
    }