Kosuke-Yamada
commited on
Commit
•
6c07a79
1
Parent(s):
75895e7
modify how to import
Browse files- ner-wikipedia-dataset.py +25 -45
ner-wikipedia-dataset.py
CHANGED
@@ -4,19 +4,7 @@ import json
|
|
4 |
import random
|
5 |
from typing import Generator
|
6 |
|
7 |
-
|
8 |
-
BuilderConfig,
|
9 |
-
DatasetInfo,
|
10 |
-
DownloadManager,
|
11 |
-
Features,
|
12 |
-
GeneratorBasedBuilder,
|
13 |
-
Sequence,
|
14 |
-
Split,
|
15 |
-
SplitGenerator,
|
16 |
-
Value,
|
17 |
-
Version,
|
18 |
-
)
|
19 |
-
from datasets.data_files import DataFilesDict
|
20 |
|
21 |
_CITATION = """
|
22 |
@inproceedings{omi-2021-wikipedia,
|
@@ -33,14 +21,14 @@ _LICENSE = "CC-BY-SA 3.0"
|
|
33 |
_URL = "https://raw.githubusercontent.com/stockmarkteam/ner-wikipedia-dataset/main/ner.json"
|
34 |
|
35 |
|
36 |
-
class NerWikipediaDatasetConfig(BuilderConfig):
|
37 |
def __init__(
|
38 |
self,
|
39 |
name: str = "default",
|
40 |
-
version: Version | str | None = Version("0.0.0"),
|
41 |
data_dir: str | None = None,
|
42 |
-
data_files: DataFilesDict | None = None,
|
43 |
-
description: str | None =
|
44 |
shuffle: bool = True,
|
45 |
seed: int = 42,
|
46 |
train_ratio: float = 0.8,
|
@@ -59,29 +47,23 @@ class NerWikipediaDatasetConfig(BuilderConfig):
|
|
59 |
self.validation_ratio = validation_ratio
|
60 |
|
61 |
|
62 |
-
class NerWikipediaDataset(GeneratorBasedBuilder):
|
63 |
BUILDER_CONFIG_CLASS = NerWikipediaDatasetConfig
|
64 |
|
65 |
-
|
66 |
-
|
67 |
-
name="ner-wikipedia-dataset",
|
68 |
-
version=Version("2.0.0"),
|
69 |
description=_DESCRIPTION,
|
70 |
-
|
71 |
-
]
|
72 |
-
|
73 |
-
def _info(self) -> DatasetInfo:
|
74 |
-
return DatasetInfo(
|
75 |
-
description=_DESCRIPTION,
|
76 |
-
features=Features(
|
77 |
{
|
78 |
-
"curid": Value("string"),
|
79 |
-
"text": Value("string"),
|
80 |
"entities": [
|
81 |
{
|
82 |
-
"name": Value("string"),
|
83 |
-
"span": Sequence(
|
84 |
-
|
|
|
|
|
85 |
}
|
86 |
],
|
87 |
}
|
@@ -92,8 +74,8 @@ class NerWikipediaDataset(GeneratorBasedBuilder):
|
|
92 |
)
|
93 |
|
94 |
def _split_generators(
|
95 |
-
self, dl_manager: DownloadManager
|
96 |
-
) -> list[SplitGenerator]:
|
97 |
dataset_dir = str(dl_manager.download_and_extract(_URL))
|
98 |
with open(dataset_dir, "r", encoding="utf-8") as f:
|
99 |
data = json.load(f)
|
@@ -106,21 +88,19 @@ class NerWikipediaDataset(GeneratorBasedBuilder):
|
|
106 |
num_train_data = int(num_data * self.config.train_ratio)
|
107 |
num_validation_data = int(num_data * self.config.validation_ratio)
|
108 |
train_data = data[:num_train_data]
|
109 |
-
validation_data = data[
|
110 |
-
num_train_data : num_train_data + num_validation_data
|
111 |
-
]
|
112 |
test_data = data[num_train_data + num_validation_data :]
|
113 |
return [
|
114 |
-
SplitGenerator(
|
115 |
-
name=Split.TRAIN,
|
116 |
gen_kwargs={"data": train_data},
|
117 |
),
|
118 |
-
SplitGenerator(
|
119 |
-
name=Split.VALIDATION,
|
120 |
gen_kwargs={"data": validation_data},
|
121 |
),
|
122 |
-
SplitGenerator(
|
123 |
-
name=Split.TEST,
|
124 |
gen_kwargs={"data": test_data},
|
125 |
),
|
126 |
]
|
|
|
4 |
import random
|
5 |
from typing import Generator
|
6 |
|
7 |
+
import datasets
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
|
9 |
_CITATION = """
|
10 |
@inproceedings{omi-2021-wikipedia,
|
|
|
21 |
_URL = "https://raw.githubusercontent.com/stockmarkteam/ner-wikipedia-dataset/main/ner.json"
|
22 |
|
23 |
|
24 |
+
class NerWikipediaDatasetConfig(datasets.BuilderConfig):
|
25 |
def __init__(
|
26 |
self,
|
27 |
name: str = "default",
|
28 |
+
version: datasets.Version | str | None = datasets.Version("0.0.0"),
|
29 |
data_dir: str | None = None,
|
30 |
+
data_files: datasets.data_files.DataFilesDict | None = None,
|
31 |
+
description: str | None = _DESCRIPTION,
|
32 |
shuffle: bool = True,
|
33 |
seed: int = 42,
|
34 |
train_ratio: float = 0.8,
|
|
|
47 |
self.validation_ratio = validation_ratio
|
48 |
|
49 |
|
50 |
+
class NerWikipediaDataset(datasets.GeneratorBasedBuilder):
|
51 |
BUILDER_CONFIG_CLASS = NerWikipediaDatasetConfig
|
52 |
|
53 |
+
def _info(self) -> datasets.DatasetInfo:
|
54 |
+
return datasets.DatasetInfo(
|
|
|
|
|
55 |
description=_DESCRIPTION,
|
56 |
+
features=datasets.Features(
|
|
|
|
|
|
|
|
|
|
|
|
|
57 |
{
|
58 |
+
"curid": datasets.Value("string"),
|
59 |
+
"text": datasets.Value("string"),
|
60 |
"entities": [
|
61 |
{
|
62 |
+
"name": datasets.Value("string"),
|
63 |
+
"span": datasets.Sequence(
|
64 |
+
datasets.Value("int64"), length=2
|
65 |
+
),
|
66 |
+
"type": datasets.Value("string"),
|
67 |
}
|
68 |
],
|
69 |
}
|
|
|
74 |
)
|
75 |
|
76 |
def _split_generators(
|
77 |
+
self, dl_manager: datasets.DownloadManager
|
78 |
+
) -> list[datasets.SplitGenerator]:
|
79 |
dataset_dir = str(dl_manager.download_and_extract(_URL))
|
80 |
with open(dataset_dir, "r", encoding="utf-8") as f:
|
81 |
data = json.load(f)
|
|
|
88 |
num_train_data = int(num_data * self.config.train_ratio)
|
89 |
num_validation_data = int(num_data * self.config.validation_ratio)
|
90 |
train_data = data[:num_train_data]
|
91 |
+
validation_data = data[num_train_data : num_train_data + num_validation_data]
|
|
|
|
|
92 |
test_data = data[num_train_data + num_validation_data :]
|
93 |
return [
|
94 |
+
datasets.SplitGenerator(
|
95 |
+
name=datasets.Split.TRAIN,
|
96 |
gen_kwargs={"data": train_data},
|
97 |
),
|
98 |
+
datasets.SplitGenerator(
|
99 |
+
name=datasets.Split.VALIDATION,
|
100 |
gen_kwargs={"data": validation_data},
|
101 |
),
|
102 |
+
datasets.SplitGenerator(
|
103 |
+
name=datasets.Split.TEST,
|
104 |
gen_kwargs={"data": test_data},
|
105 |
),
|
106 |
]
|