ciral.py

# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.


import csv
import json
import os

import datasets
from collections import defaultdict


_CITATION = ""

languages = {'yoruba':'yo', 
             'hausa':'ha',
             'swahili':'sw',
             'somali':'so'}

# TODO: Add description of the dataset here
# You can copy an official description
_DESCRIPTION = """\
This dataset consists of the queries and relevance judgements in the CIRAL test collection. 
"""

_HOMEPAGE = ""

_LICENSE = ""

_URLS = {
    lang: {
        'dev': [
            f'https://huggingface.co/datasets/CIRAL/ciral/resolve/main/ciral-{lang}/topics/topics.ciral-v1.0-{lang_code}-dev.tsv',
            f'https://huggingface.co/datasets/CIRAL/ciral/resolve/main/ciral-{lang}/qrels/qrels.ciral-v1.0-{lang_code}-dev.tsv'
        ],
        'testA':[
            f'https://huggingface.co/datasets/CIRAL/ciral/resolve/main/ciral-{lang}/topics/topics.ciral-v1.0-{lang_code}-test-a.tsv',
            f'https://huggingface.co/datasets/CIRAL/ciral/resolve/main/ciral-{lang}/qrels/qrels.ciral-v1.0-{lang_code}-test-a.tsv',
            f'https://huggingface.co/datasets/CIRAL/ciral/resolve/main/ciral-{lang}/qrels/qrels.ciral-v1.0-{lang_code}-test-a-pools.tsv',
        ],
        'testB':[
            f'https://huggingface.co/datasets/CIRAL/ciral/resolve/main/ciral-{lang}/topics/topics.ciral-v1.0-{lang_code}-test-b.tsv',
            f'https://huggingface.co/datasets/CIRAL/ciral/resolve/main/ciral-{lang}/qrels/qrels.ciral-v1.0-{lang_code}-test-b.tsv',
        ]
    } for lang, lang_code in languages.items()
}


def load_queries(_file):
    if _file is None:
        return []

    queries = {}
    with open(_file, encoding="utf-8") as query_file:
        for line in query_file:
            line = line.strip()
            id, query = (line.split('\t')) if len(line.split('\t')) == 2 else ("", "")
            queries[id] = query
    return queries

def load_qrels(_file):
    if _file is None:
        return None

    qrels = defaultdict(dict)
    with open(_file, encoding="utf-8") as qrel_file:
        for line in qrel_file:
            line = line.strip()
            qid, _, docid, rel = (line.split('\t')) if len(line.split('\t')) == 4 else ("", "", "",False)
            qrels[qid][docid] = int(rel)
    #print(qrels)
    return qrels



class CIRAL(datasets.GeneratorBasedBuilder):
    #VERSION = datasets.Version("1.1.0")

    BUILDER_CONFIGS = [
        datasets.BuilderConfig(
            name=lang,
            version=datasets.Version("1.1.0"), 
            description=f"CIRAL data for {lang}.")  for lang in languages.keys()
    ]

    def _info(self):
        features = datasets.Features(
            {
                "query_id": datasets.Value("string"),
                "query": datasets.Value("string"),
                # "judgements": [{
                #     "docid": datasets.Value("string"),
                #     "judgement": datasets.Value("string"),
                #     "text": datasets.Value("string")}]
                "positive_passages": [{
                    'docid': datasets.Value("string"),
                    'text': datasets.Value("string"),
                }],
                "negative_passages": [{
                    "docid": datasets.Value("string"),
                    "text": datasets.Value("string")
                }],
        
                "pools_positive_passages": [{
                    'docid': datasets.Value("string"),
                    'text': datasets.Value("string"),
                }],
                "pools_negative_passages": [{
                    "docid": datasets.Value("string"),
                    "text": datasets.Value("string")}]
            }
        )

        return datasets.DatasetInfo(
            # This is the description that will appear on the datasets page.
            description=_DESCRIPTION,
            # This defines the different columns of the dataset and their types
            features=features,  
            supervised_keys=None,
            homepage=_HOMEPAGE,
            license=_LICENSE,
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        
        lang = self.config.name
        downloaded_files = dl_manager.download_and_extract(_URLS[lang])
        return [
            datasets.SplitGenerator(
                name='dev',
                gen_kwargs={
                    'filepaths': downloaded_files['dev'],
                },
            ),
            datasets.SplitGenerator(
                name='testA',
                gen_kwargs={
                    'filepaths': downloaded_files['testA'],
                },
            ),
            datasets.SplitGenerator(
                name='testB',
                gen_kwargs={
                    'filepaths': downloaded_files['testB'],
                },
            ),
        ]

    # method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
    def _generate_examples(self, filepaths):
        lang = self.config.name
        corpus = datasets.load_dataset('ciral/ciral-corpus', lang)['train']
        docid2doc = {doc['docid']: doc['text'] for doc in corpus}

        query_file, qrel_file, pools_file = (filepaths) if len(filepaths) == 3 else (filepaths[0], filepaths[1], None)
        queries = load_queries(query_file)
        shallow_qrels = load_qrels(qrel_file)
        pools_qrels = load_qrels(pools_file)
        
        for query_id in queries:
            positive_docids = [docid for docid, judgement in shallow_qrels[query_id].items() if judgement==1]
            negative_docids = [docid for docid, judgement in shallow_qrels[query_id].items() if judgement==0]

            pools_positive_docids = [docid for docid, judgement in pools_qrels[query_id].items() if judgement==1] if pools_qrels is not None else []
            pools_negative_docids = [docid for docid, judgement in pools_qrels[query_id].items() if judgement==0] if pools_qrels is not None else []

            data = {}
            data['query_id'] = query_id
            data['query'] = queries[query_id]
            data['positive_passages'] = [{
                'docid': docid,
                'text': docid2doc[docid]
            } for docid in positive_docids if docid in docid2doc]
            data['negative_passages'] = [{
                'docid': docid,
                'text': docid2doc[docid]
            } for docid in negative_docids if docid in docid2doc]

            data['pools_positive_passages'] = [{
                'docid': docid,
                'text': docid2doc[docid]
            } for docid in pools_positive_docids if docid in docid2doc]
            data['pools_negative_passages'] = [{
                'docid': docid,
                'text': docid2doc[docid]
            } for docid in pools_negative_docids if docid in docid2doc]

            yield query_id, data