cannabis_tests.py

"""
Cannabis Tests
Copyright (c) 2022 Cannlytics

Authors:
    Keegan Skeate <https://github.com/keeganskeate>
    Candace O'Sullivan-Sutherland <https://github.com/candy-o>
Created: 9/10/2022
Updated: 9/10/2022
License: <https://github.com/cannlytics/cannlytics/blob/main/LICENSE>
"""
import json
import datasets # pip install datasets

# Citation for using the dataset.
CANNABIS_TESTS_CITATION = """\
@inproceedings{cannlytics2022cannabis_tests,
  author    = {Skeate, Keegan and O'Sullivan-Sutherland, Candace},
  title     = {Cannabis Tests: Curated Cannabis Lab Test Results},
  booktitle = {Cannabis Data Science},
  month     = {September},
  year      = {2022},
  address   = {United States of America},
  publisher = {Cannlytics}
}
"""
CANNABIS_TESTS_DESCRIPTION = """\
Cannabis lab test results (https://cannlytics.com/data/tests) is a
dataset of curated cannabis lab test results.
"""
CANNABIS_TESTS_URL = 'https://huggingface.co/datasets/cannlytics/cannabis_tests'

# Raw Garden constants.
RAWGARDEN_DATA_URL = 'https://cannlytics.page.link/rawgarden'
RAWGARDEN_URL = 'https://github.com/cannlytics/cannlytics/tree/main/ai/curation/get_rawgarden_data'
RAWGARDEN_DESCRIPTION = """\
Raw Garden lab test results (https://cannlytics.com/data/tests) is a
dataset of curated cannabis lab test results from Raw Garden, a large
cannabis processor in California.
"""
RAWGARDEN_FEATURES = []

# MCR Labs constants.
MCRLABS_DATA_URL = 'https://cannlytics.page.link/mcrlabs'
MCRLABS_URL = 'https://github.com/cannlytics/cannlytics/tree/main/ai/curation/get_mcr_labs_data'
MCRLABS_DESCRIPTION = """\
MCR Labs lab test results (https://cannlytics.com/data/tests) is a
dataset of curated cannabis lab test results from MCR Labs, a lab that
tests cannabis in Massachusetts.
"""
MCRLABS_FEATURES = []

# PSI Labs constants.
PSILABS_DATA_URL = 'https://cannlytics.page.link/psilabs'
PSILABS_URL = 'https://github.com/cannlytics/cannlytics/tree/main/ai/curation/get_psi_labs_data'
PSILABS_DESCRIPTION = """\
PSI Labs lab test results (https://cannlytics.com/data/tests) is a
dataset of curated cannabis lab test results from PSI Labs, a lab that
tests cannabis in Michigan.
"""
PSILABS_FEATURES = []

# SC Labs constants.
SCLABS_DATA_URL = 'https://cannlytics.page.link/sclabs'
SCLABS_URL = 'https://github.com/cannlytics/cannlytics/tree/main/ai/curation/get_sc_labs_data'
SCLABS_DESCRIPTION = """\
SC Labs lab test results (https://cannlytics.com/data/tests) is a
dataset of curated cannabis lab test results from SC Labs, a lab that
tests cannabis in California.
"""
SCLABS_FEATURES = []


class CannabisTestsConfig(datasets.BuilderConfig):
    """BuilderConfig for Cannabis Tests."""

    def __init__(
            self,
            features,
            data_url,
            citation,
            url,
            label_classes=('False', 'True'),
            **kwargs
        ):
        """BuilderConfig for Cannabis Tests.
        Args:
        features (list[str]): A list of the features that will appear in
            the feature dict. Should not include "label".
        data_url (str): A URL to download the zip file from.
        citation (str): A citation for the data set.
        url (str): A URL for information about the data set.
        label_classes (list[str]): The list of classes for the label if
            the label is present as a string. Non-string labels will be
            cast to either 'False' or 'True'.
        **kwargs: Keyword arguments forwarded to super.
        """
        super(CannabisTestsConfig, self).__init__(version=datasets.Version("1.0.2"), **kwargs)
        self.features = features
        self.label_classes = label_classes
        self.data_url = data_url
        self.citation = citation
        self.url = url


class CannabisTests(datasets.GeneratorBasedBuilder):
    """The Cannabis Tests dataset."""

    VERSION = datasets.Version('1.0.0')

    BUILDER_CONFIGS = [
        CannabisTestsConfig(
            name='rawgarden',
            description=RAWGARDEN_DESCRIPTION,
            features=RAWGARDEN_FEATURES,
            data_url=RAWGARDEN_DATA_URL,
            citation=CANNABIS_TESTS_CITATION,
            url=RAWGARDEN_URL,
        ),
        CannabisTestsConfig(
            name='mcrlabs',
            description=MCRLABS_DESCRIPTION,
            features=MCRLABS_FEATURES,
            data_url=MCRLABS_DATA_URL,
            citation=CANNABIS_TESTS_CITATION,
            url=MCRLABS_URL,
        ),
        CannabisTestsConfig(
            name='psilabs',
            description=PSILABS_DESCRIPTION,
            features=PSILABS_FEATURES,
            data_url=PSILABS_DATA_URL,
            citation=CANNABIS_TESTS_CITATION,
            url=PSILABS_URL,
        ),
        CannabisTestsConfig(
            name='sclabs',
            description=SCLABS_DESCRIPTION,
            features=SCLABS_FEATURES,
            data_url=SCLABS_DATA_URL,
            citation=CANNABIS_TESTS_CITATION,
            url=SCLABS_URL,
        ),
    ]

    DEFAULT_CONFIG_NAME = 'rawgarden'

    def _info(self):
        features = {feature: datasets.Value('string') for feature in self.config.features}

        # TODO: Define all numeric features.
        # features['span1_index'] = datasets.Value('int32')

        # TODO: Define all image features.

        # TODO: Define all sequence features.
        # features["answers"] = datasets.features.Sequence(
        #     {"text": datasets.Value("string"), "answer_start": datasets.Value("int32"),}
        # )

        return datasets.DatasetInfo(
            citation=CANNABIS_TESTS_CITATION,
            description=CANNABIS_TESTS_DESCRIPTION,
            features=features,
            homepage=CANNABIS_TESTS_URL,
            supervised_keys=None,
        )    

    def _generate_examples(self, filepath):
        """This function returns the examples in raw (text) form."""
        with open(filepath,  encoding='utf-8') as f:
            for line in f:
                row = json.loads(line)
                product_name = row.get('product_name', '').strip()
                for i, result in enumerate(row['results']):
                    _id = str(i)
                    yield _id, {
                        'product_name': product_name,
                        'analyte': result.get('key', ''),
                        'value': result.get('value', 0),
                    }