nyt_ingredients.py

"""New York Times Ingredient Phrase Tagger Dataset"""


import datasets


logger = datasets.logging.get_logger(__name__)


_CITATION = """\
@misc{nytimesTaggedIngredients,
	author = {Erica Greene and Adam Mckaig},
	title = {{O}ur {T}agged {I}ngredients {D}ata is {N}ow on {G}it{H}ub --- archive.nytimes.com},
	howpublished = {\\url{https://archive.nytimes.com/open.blogs.nytimes.com/2016/04/27/structured-ingredients-data-tagging/}},
	year = {},
	note = {[Accessed 03-10-2023]},
}
"""


_DESCRIPTION = """\
New York Times Ingredient Phrase Tagger Dataset
We use a conditional random field model (CRF) to extract tags from labelled training data, which was tagged by human news assistants. 
e wrote about our approach on the [New York Times Open blog](http://open.blogs.nytimes.com/2015/04/09/extracting-structured-data-from-recipes-using-conditional-random-fields/).
This repo contains scripts to extract the Quantity, Unit, Name, and Comments from unstructured ingredient phrases. 
We use it on Cooking to format incoming recipes. Given the following input:

```
1 pound carrots, young ones if possible
Kosher salt, to taste
2 tablespoons sherry vinegar
2 tablespoons honey
2 tablespoons extra-virgin olive oil
1 medium-size shallot, peeled and finely diced
1/2 teaspoon fresh thyme leaves, finely chopped
Black pepper, to taste
```
"""

_URL = "https://github.com/nytimes/ingredient-phrase-tagger"


_URLS = {
    "train": "https://huggingface.co/datasets/napsternxg/nyt_ingredients/resolve/main/nyt-ingredients.crf.jsonl"
}

import json


class NYTIngredientsConfig(datasets.BuilderConfig):
    """The NYT Ingredients Dataset."""

    def __init__(self, **kwargs):
        """BuilderConfig for NYT Ingredients.
        Args:
          **kwargs: keyword arguments forwarded to super.
        """
        super(NYTIngredientsConfig, self).__init__(**kwargs)


class NYTIngredients(datasets.GeneratorBasedBuilder):
    """The NYT Ingredients Dataset."""

    BUILDER_CONFIGS = [
        NYTIngredientsConfig(
            name="nyt_ingredients",
            version=datasets.Version("1.0.0"),
            description="The NYT Ingredients Dataset",
        ),
    ]

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "input": datasets.Value("string"),
                    "display_input": datasets.Value("string"),
                    "tokens": datasets.Sequence(datasets.Value("string")),
                    "index": datasets.Sequence(datasets.Value("string")),
                    "lengthGroup": datasets.Sequence(datasets.Value("string")),
                    "isCapitalized": datasets.Sequence(
                        datasets.features.ClassLabel(
                            names=[
                                "NoCAP",
                                "YesCAP"
                            ]
                        )
                    ),
                    "insideParenthesis": datasets.Sequence(
                        datasets.features.ClassLabel(
                            names=[
                                "NoPAREN",
                                "YesPAREN",
                            ]
                        )
                    ),
                    "label": datasets.Sequence(
                        datasets.features.ClassLabel(
                            names=[
                                "O",
                                "B-COMMENT",
                                "I-COMMENT",
                                "B-NAME",
                                "I-NAME",
                                "B-RANGE_END",
                                "I-RANGE_END",
                                "B-QTY",
                                "I-QTY",
                                "B-UNIT",
                                "I-UNIT",
                            ]
                        )
                    ),
                }
            ),
            supervised_keys=None,
            homepage="https://github.com/nytimes/ingredient-phrase-tagger",
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        downloaded_files = dl_manager.download_and_extract(_URLS)
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={"filepath": downloaded_files["train"]},
            ),
        ]

    def _generate_examples(self, filepath):
        logger.info("⏳ Generating examples from = %s", filepath)
        with open(filepath, encoding="utf-8") as fp:
            for i, line in enumerate(fp):
                yield i, json.loads(line)