|
|
|
'''DiscEvalMT: DiscEvalMT: Contrastive test sets for the evaluation of discourse in machine translation (v2)''' |
|
|
|
import json |
|
import datasets |
|
|
|
logger = datasets.logging.get_logger(__name__) |
|
|
|
_CITATION = '''\ |
|
@inproceedings{bawden-etal-2018-evaluating, |
|
title = "Evaluating Discourse Phenomena in Neural Machine Translation", |
|
author = "Bawden, Rachel and Sennrich, Rico and Birch, Alexandra and Haddow, Barry", |
|
booktitle = {{Proceedings of the 2018 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers)}}, |
|
month = jun, |
|
year = "2018", |
|
address = "New Orleans, Louisiana", |
|
publisher = "Association for Computational Linguistics", |
|
url = "https://www.aclweb.org/anthology/N18-1118", |
|
doi = "10.18653/v1/N18-1118", |
|
pages = "1304--1313" |
|
} |
|
''' |
|
|
|
_DESCRIPTION = '''\ |
|
English-French hand-crafted contrastive test set to test anaphora and lexical choice |
|
''' |
|
|
|
_HOMEPAGE='https://github.com/rbawden/discourse-mt-test-sets/tree/master' |
|
|
|
_LICENSE = 'CC-BY-SA-4.0' |
|
|
|
_URLS = { |
|
'test-lexical_choice': 'lexical_choice/eval.jsonl', |
|
'test-anaphora': 'anaphora/eval.jsonl' |
|
} |
|
|
|
|
|
class DiscEvalMTConfig(datasets.BuilderConfig): |
|
'''BuilderConfig for DiscEvalMT.''' |
|
|
|
|
|
def __init__(self, evaltype: str, **kwargs): |
|
"""BuilderConfig for DiscEvalMT. |
|
|
|
Args: |
|
**kwargs: keyword arguments forwarded to super. |
|
""" |
|
self.evaltype = evaltype |
|
if evaltype not in ['anaphora', 'lexical_choice']: |
|
raise ValueError("Invalid evaltype: %s. You must choose between 'anaphora' and 'lexical_choice' " % evaltype) |
|
super(DiscEvalMTConfig, self).__init__(**kwargs) |
|
|
|
|
|
|
|
|
|
class DiscEvalMT(datasets.GeneratorBasedBuilder): |
|
'''DiscEvalMT: English-French contrastive test set for 2 discourse phenomena (anaphora and lexical cohesion)''' |
|
|
|
BUILDER_CONFIG_CLASS = DiscEvalMTConfig |
|
BUILDER_CONFIGS = [ |
|
DiscEvalMTConfig( |
|
evaltype='anaphora', |
|
name='anaphora', |
|
version=datasets.Version('2.0.0'), |
|
), |
|
DiscEvalMTConfig( |
|
evaltype='lexical_choice', |
|
name='lexical_choice', |
|
version=datasets.Version('2.0.0'), |
|
), |
|
] |
|
|
|
def _info(self): |
|
return datasets.DatasetInfo( |
|
description=_DESCRIPTION, |
|
features=datasets.Features({ |
|
"split": datasets.Value("string"), |
|
"ex_num": datasets.Value("int64"), |
|
"type": datasets.Value("string"), |
|
"context_src": datasets.Value("string"), |
|
"current_src": datasets.Value("string"), |
|
"context_trg": datasets.Value("string"), |
|
"current_trg": datasets.Value("string"), |
|
"contrastive_context_trg": datasets.Value("string"), |
|
"contrastive_current_trg": datasets.Value("string"), |
|
"correct_or_semicorrect": datasets.Value("string")}), |
|
supervised_keys=None, |
|
homepage=_HOMEPAGE, |
|
citation=_CITATION, |
|
license=_LICENSE, |
|
) |
|
|
|
def _split_generators(self, dl_manager): |
|
downloaded_files = dl_manager.download_and_extract(_URLS) |
|
|
|
return datasets.SplitGenerator(name="test", gen_kwargs={'filepath': downloaded_files['test-' + self.config.evaltype]}), |
|
|
|
|
|
def _generate_examples(self, filepath): |
|
logger.info("generating examples from = %s", filepath) |
|
key = 0 |
|
with open(filepath, encoding="utf-8") as f: |
|
for i, line in enumerate(f): |
|
|
|
example = json.loads(line) |
|
yield i, example |
|
|
