Commit
•
3e0e462
1
Parent(s):
67f64b8
Convert dataset to Parquet (#7)
Browse files- Convert dataset to Parquet (ec6839eed447dfffbd0bfb7551701b7dcadd44aa)
- Add 'snli' config data files (cdf18d97fcea5bf4a6d7a6e9365ea20d4024eaae)
- Add 'xnli' config data files (17289f0bafe71b2cc393b7d1530cf24db6f18f47)
- Delete loading script (a4fdb7b517f363a92fd8cf5a447301115c6a4d81)
- Delete data file (a2a6a1a066884fb48ec27a1e66152f0184233c6e)
- README.md +25 -10
- kor_nli.py +0 -121
- data.zip → multi_nli/train-00000-of-00001.parquet +2 -2
- snli/train-00000-of-00001.parquet +3 -0
- xnli/test-00000-of-00001.parquet +3 -0
- xnli/validation-00000-of-00001.parquet +3 -0
README.md
CHANGED
@@ -39,10 +39,10 @@ dataset_info:
|
|
39 |
'2': contradiction
|
40 |
splits:
|
41 |
- name: train
|
42 |
-
num_bytes:
|
43 |
num_examples: 392702
|
44 |
-
download_size:
|
45 |
-
dataset_size:
|
46 |
- config_name: snli
|
47 |
features:
|
48 |
- name: premise
|
@@ -58,10 +58,10 @@ dataset_info:
|
|
58 |
'2': contradiction
|
59 |
splits:
|
60 |
- name: train
|
61 |
-
num_bytes:
|
62 |
num_examples: 550152
|
63 |
-
download_size:
|
64 |
-
dataset_size:
|
65 |
- config_name: xnli
|
66 |
features:
|
67 |
- name: premise
|
@@ -77,13 +77,28 @@ dataset_info:
|
|
77 |
'2': contradiction
|
78 |
splits:
|
79 |
- name: validation
|
80 |
-
num_bytes:
|
81 |
num_examples: 2490
|
82 |
- name: test
|
83 |
-
num_bytes:
|
84 |
num_examples: 5010
|
85 |
-
download_size:
|
86 |
-
dataset_size:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
87 |
---
|
88 |
|
89 |
# Dataset Card for "kor_nli"
|
|
|
39 |
'2': contradiction
|
40 |
splits:
|
41 |
- name: train
|
42 |
+
num_bytes: 84728887
|
43 |
num_examples: 392702
|
44 |
+
download_size: 54693610
|
45 |
+
dataset_size: 84728887
|
46 |
- config_name: snli
|
47 |
features:
|
48 |
- name: premise
|
|
|
58 |
'2': contradiction
|
59 |
splits:
|
60 |
- name: train
|
61 |
+
num_bytes: 80136649
|
62 |
num_examples: 550152
|
63 |
+
download_size: 22015955
|
64 |
+
dataset_size: 80136649
|
65 |
- config_name: xnli
|
66 |
features:
|
67 |
- name: premise
|
|
|
77 |
'2': contradiction
|
78 |
splits:
|
79 |
- name: validation
|
80 |
+
num_bytes: 518822
|
81 |
num_examples: 2490
|
82 |
- name: test
|
83 |
+
num_bytes: 1047429
|
84 |
num_examples: 5010
|
85 |
+
download_size: 529321
|
86 |
+
dataset_size: 1566251
|
87 |
+
configs:
|
88 |
+
- config_name: multi_nli
|
89 |
+
data_files:
|
90 |
+
- split: train
|
91 |
+
path: multi_nli/train-*
|
92 |
+
- config_name: snli
|
93 |
+
data_files:
|
94 |
+
- split: train
|
95 |
+
path: snli/train-*
|
96 |
+
- config_name: xnli
|
97 |
+
data_files:
|
98 |
+
- split: validation
|
99 |
+
path: xnli/validation-*
|
100 |
+
- split: test
|
101 |
+
path: xnli/test-*
|
102 |
---
|
103 |
|
104 |
# Dataset Card for "kor_nli"
|
kor_nli.py
DELETED
@@ -1,121 +0,0 @@
|
|
1 |
-
"""TODO(kor_nli): Add a description here."""
|
2 |
-
|
3 |
-
|
4 |
-
import os
|
5 |
-
|
6 |
-
import datasets
|
7 |
-
|
8 |
-
|
9 |
-
# TODO(kor_nli): BibTeX citation
|
10 |
-
_CITATION = """\
|
11 |
-
@article{ham2020kornli,
|
12 |
-
title={KorNLI and KorSTS: New Benchmark Datasets for Korean Natural Language Understanding},
|
13 |
-
author={Ham, Jiyeon and Choe, Yo Joong and Park, Kyubyong and Choi, Ilji and Soh, Hyungjoon},
|
14 |
-
journal={arXiv preprint arXiv:2004.03289},
|
15 |
-
year={2020}
|
16 |
-
}
|
17 |
-
"""
|
18 |
-
|
19 |
-
# TODO(kor_nli):
|
20 |
-
_DESCRIPTION = """ Korean Natural Language Inference datasets
|
21 |
-
"""
|
22 |
-
_URL = "data.zip"
|
23 |
-
|
24 |
-
|
25 |
-
class KorNLIConfig(datasets.BuilderConfig):
|
26 |
-
"""BuilderConfig for KorNLI."""
|
27 |
-
|
28 |
-
def __init__(self, **kwargs):
|
29 |
-
"""BuilderConfig for KorNLI.
|
30 |
-
|
31 |
-
Args:
|
32 |
-
|
33 |
-
**kwargs: keyword arguments forwarded to super.
|
34 |
-
"""
|
35 |
-
# Version 1.1.0 remove empty document and summary strings.
|
36 |
-
super(KorNLIConfig, self).__init__(version=datasets.Version("1.0.0"), **kwargs)
|
37 |
-
|
38 |
-
|
39 |
-
class KorNli(datasets.GeneratorBasedBuilder):
|
40 |
-
"""TODO(kor_nli): Short description of my dataset."""
|
41 |
-
|
42 |
-
# TODO(kor_nli): Set up version.
|
43 |
-
VERSION = datasets.Version("1.0.0")
|
44 |
-
BUILDER_CONFIGS = [
|
45 |
-
KorNLIConfig(name="multi_nli", description="Korean multi NLI datasets"),
|
46 |
-
KorNLIConfig(name="snli", description="Korean SNLI dataset"),
|
47 |
-
KorNLIConfig(name="xnli", description="Korean XNLI dataset"),
|
48 |
-
]
|
49 |
-
|
50 |
-
def _info(self):
|
51 |
-
# TODO(kor_nli): Specifies the datasets.DatasetInfo object
|
52 |
-
return datasets.DatasetInfo(
|
53 |
-
# This is the description that will appear on the datasets page.
|
54 |
-
description=_DESCRIPTION,
|
55 |
-
# datasets.features.FeatureConnectors
|
56 |
-
features=datasets.Features(
|
57 |
-
{
|
58 |
-
# These are the features of your dataset like images, labels ...
|
59 |
-
"premise": datasets.Value("string"),
|
60 |
-
"hypothesis": datasets.Value("string"),
|
61 |
-
"label": datasets.ClassLabel(names=["entailment", "neutral", "contradiction"]),
|
62 |
-
}
|
63 |
-
),
|
64 |
-
# If there's a common (input, target) tuple from the features,
|
65 |
-
# specify them here. They'll be used if as_supervised=True in
|
66 |
-
# builder.as_dataset.
|
67 |
-
supervised_keys=None,
|
68 |
-
# Homepage of the dataset for documentation
|
69 |
-
homepage="https://github.com/kakaobrain/KorNLUDatasets",
|
70 |
-
citation=_CITATION,
|
71 |
-
)
|
72 |
-
|
73 |
-
def _split_generators(self, dl_manager):
|
74 |
-
"""Returns SplitGenerators."""
|
75 |
-
# TODO(kor_nli): Downloads the data and defines the splits
|
76 |
-
# dl_manager is a datasets.download.DownloadManager that can be used to
|
77 |
-
# download and extract URLs
|
78 |
-
dl_dir = dl_manager.download_and_extract(_URL)
|
79 |
-
dl_dir = os.path.join(dl_dir, "KorNLI")
|
80 |
-
if self.config.name == "multi_nli":
|
81 |
-
return [
|
82 |
-
datasets.SplitGenerator(
|
83 |
-
name=datasets.Split.TRAIN,
|
84 |
-
# These kwargs will be passed to _generate_examples
|
85 |
-
gen_kwargs={"filepath": os.path.join(dl_dir, "multinli.train.ko.tsv")},
|
86 |
-
),
|
87 |
-
]
|
88 |
-
elif self.config.name == "snli":
|
89 |
-
return [
|
90 |
-
datasets.SplitGenerator(
|
91 |
-
name=datasets.Split.TRAIN,
|
92 |
-
# These kwargs will be passed to _generate_examples
|
93 |
-
gen_kwargs={"filepath": os.path.join(dl_dir, "snli_1.0_train.ko.tsv")},
|
94 |
-
),
|
95 |
-
]
|
96 |
-
else:
|
97 |
-
return [
|
98 |
-
datasets.SplitGenerator(
|
99 |
-
name=datasets.Split.VALIDATION,
|
100 |
-
# These kwargs will be passed to _generate_examples
|
101 |
-
gen_kwargs={"filepath": os.path.join(dl_dir, "xnli.dev.ko.tsv")},
|
102 |
-
),
|
103 |
-
datasets.SplitGenerator(
|
104 |
-
name=datasets.Split.TEST,
|
105 |
-
# These kwargs will be passed to _generate_examples
|
106 |
-
gen_kwargs={"filepath": os.path.join(dl_dir, "xnli.test.ko.tsv")},
|
107 |
-
),
|
108 |
-
]
|
109 |
-
|
110 |
-
def _generate_examples(self, filepath):
|
111 |
-
"""Yields examples."""
|
112 |
-
# TODO(kor_nli): Yields (key, example) tuples from the dataset
|
113 |
-
with open(filepath, encoding="utf-8") as f:
|
114 |
-
next(f) # skip headers
|
115 |
-
columns = ("premise", "hypothesis", "label")
|
116 |
-
for id_, row in enumerate(f):
|
117 |
-
row = row.strip().split("\t")
|
118 |
-
if len(row) != 3:
|
119 |
-
continue
|
120 |
-
row = dict(zip(columns, row))
|
121 |
-
yield id_, row
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
data.zip → multi_nli/train-00000-of-00001.parquet
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8de5f800e4c9d0fb89bd6a36a800f7ec26be2a97f50c32f2aff3c197ca155f66
|
3 |
+
size 54693610
|
snli/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dd6c2d95ae7b8979e2a89c0d05b208f38d480ca4e23f8249122e9b40116b5572
|
3 |
+
size 22015955
|
xnli/test-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2ebba3632ffac4d05a18e76629486eb88c584ae5459d15181c295341e6af659b
|
3 |
+
size 351479
|
xnli/validation-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ef903c790a16149f167367a01775aaf7e5d6d1a74deb89400eec4f3f29b4aec8
|
3 |
+
size 177842
|