Datasets:

Tasks:
Other
Modalities:
Text
Languages:
Swedish
Libraries:
Datasets
License:
sucx3_ner / sucx3_ner.py
Robin Kurtz
...
f2d5b47
# coding=utf-8
# Copyright 2020 The TensorFlow Datasets Authors and the HuggingFace Datasets Authors.
#
# 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.
# Lint as: python3
"""The SuperGLUE benchmark."""
import json
import os
import datasets
_CITATION = """\
@article{gustafson2006documentation,
title={Documentation of the Stockholm-Ume{\aa} Corpus},
author={Gustafson-Capkov{\'a}, Sofia and Hartmann, Britt},
journal={Stockholm University: Department of Linguistics},
year={2006}
}
"""
# You can copy an official description
_DESCRIPTION = """\
The dataset is a conversion of the venerable SUC 3.0 dataset into the
huggingface ecosystem. The original dataset does not contain an official
train-dev-test split, which is introduced here; the tag distribution for the
NER tags between the three splits is mostly the same.
The dataset has three different types of tagsets: manually annotated POS,
manually annotated NER, and automatically annotated NER. For the
automatically annotated NER tags, only sentences were chosen, where the
automatic and manual annotations would match (with their respective
categories).
Additionally we provide remixes of the same data with some or all sentences
being lowercased.
"""
_HOMEPAGE = "https://spraakbanken.gu.se/en/resources/suc3"
_LICENSE = "CC-BY-4.0"
# TODO: Add link to the official dataset URLs here
# The HuggingFace Datasets library doesn't host the datasets but only points to the original files.
# This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method)
# _URL = "https://huggingface.co/datasets/KBLab/suc3_1/resolve/main/data/"
_URL = "https://huggingface.co/datasets/KBLab/sucx3_ner/resolve/main/data/"
_URLS = {
"original_tags": {
"cased": "original_tags/cased.tar.gz",
"lower": "original_tags/lower.tar.gz",
"lower_mix": "original_tags/lower_mix.tar.gz"},
"simple_tags": {
"cased": "simple_tags/cased.tar.gz",
"lower": "simple_tags/lower.tar.gz",
"lower_mix": "simple_tags/lower_mix.tar.gz"}
}
_POS_LABEL_NAMES = {
'AB', 'DT', 'HA', 'HD', 'HP', 'HS', 'IE', 'IN', 'JJ', 'KN', 'MAD', 'MID',
'NN', 'PAD', 'PC', 'PL', 'PM', 'PN', 'PP', 'PS', 'RG', 'RO', 'SN', 'UO',
'VB'
}
_NER_LABEL_NAMES_ORIGINAL = {
'B-animal', 'B-event', 'B-inst', 'B-myth', 'B-other', 'B-person',
'B-place', 'B-product', 'B-work', 'I-animal', 'I-event', 'I-inst',
'I-myth', 'I-other', 'I-person', 'I-place', 'I-product', 'I-work', 'O'
}
_NER_LABEL_NAMES_SIMPLE = {
'B-EVN', 'B-LOC', 'B-MSR', 'B-OBJ', 'B-ORG', 'B-PRS', 'B-TME', 'B-WRK',
'I-EVN', 'I-LOC', 'I-MSR', 'I-OBJ', 'I-ORG', 'I-PRS', 'I-TME', 'I-WRK', 'O'
}
class SUC3Config(datasets.BuilderConfig):
"""BuilderConfig for Suc."""
def __init__(self,
ner_label_names,
description,
data_url,
**kwargs):
"""BuilderConfig for Suc.
"""
super(SUC3Config,
self).__init__(version=datasets.Version("1.0.2"), **kwargs)
self.ner_label_names = ner_label_names
self.description = description
self.data_url = data_url
class SUC3(datasets.GeneratorBasedBuilder):
"""The SuperGLUE benchmark."""
BUILDER_CONFIGS = [
SUC3Config(
name="original_cased",
ner_label_names=_NER_LABEL_NAMES_ORIGINAL,
data_url=_URLS["original_tags"]["cased"],
description="manually annotated & cased",
),
SUC3Config(
name="original_lower",
ner_label_names=_NER_LABEL_NAMES_ORIGINAL,
data_url=_URLS["original_tags"]["lower"],
description="manually annotated & lower",
),
SUC3Config(
name="original_lower_mix",
ner_label_names=_NER_LABEL_NAMES_ORIGINAL,
data_url=_URLS["original_tags"]["lower_mix"],
description="manually annotated & lower_mix",
),
SUC3Config(
name="simple_cased",
ner_label_names=_NER_LABEL_NAMES_SIMPLE,
data_url=_URLS["simple_tags"]["cased"],
description="automatically annotated & cased",
),
SUC3Config(
name="simple_lower",
ner_label_names=_NER_LABEL_NAMES_SIMPLE,
data_url=_URLS["simple_tags"]["lower"],
description="automatically annotated & lower",
),
SUC3Config(
name="simple_lower_mix",
ner_label_names=_NER_LABEL_NAMES_SIMPLE,
data_url=_URLS["simple_tags"]["lower_mix"],
description="autimatically annotated & lower_mix",
),
]
def _info(self):
features = {"id": datasets.Value("string"),
"tokens": datasets.features.Sequence(datasets.Value("string")),
"pos_tags": datasets.features.Sequence(datasets.Value("string")),
"ner_tags": datasets.features.Sequence(datasets.Value("string"))}
return datasets.DatasetInfo(
description=_DESCRIPTION + self.config.description,
features=datasets.Features(features),
homepage=_HOMEPAGE,
citation=_CITATION,
supervised_keys=None,
)
def _split_generators(self, dl_manager):
dl_dir = dl_manager.download_and_extract(_URL + self.config.data_url)
dl_dir = os.path.join(dl_dir, self.config.data_url.split("/")[-1].split(".")[0])
return [
datasets.SplitGenerator(
name=datasets.Split.TRAIN,
gen_kwargs={
"data_file": os.path.join(dl_dir, "train.jsonl"),
},
),
datasets.SplitGenerator(
name=datasets.Split.VALIDATION,
gen_kwargs={
"data_file": os.path.join(dl_dir, "dev.jsonl"),
},
),
datasets.SplitGenerator(
name=datasets.Split.TEST,
gen_kwargs={
"data_file": os.path.join(dl_dir, "test.jsonl"),
},
),
]
def _generate_examples(self, data_file):
with open(data_file, encoding="utf-8") as f:
for i, line in enumerate(f):
row = json.loads(line)
yield str(i), row