Create dsr.py
Browse files
dsr.py
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import datasets
|
2 |
+
import csv
|
3 |
+
import requests
|
4 |
+
|
5 |
+
|
6 |
+
citation='''
|
7 |
+
'''
|
8 |
+
|
9 |
+
class Config(datasets.BuilderConfig):
|
10 |
+
citation=citation
|
11 |
+
|
12 |
+
files = [
|
13 |
+
"eng.dep.covdtb",
|
14 |
+
"eng.dep.scidtb",
|
15 |
+
"eng.pdtb.pdtb",
|
16 |
+
"eng.pdtb.tedm",
|
17 |
+
"eng.rst.gum",
|
18 |
+
"eng.rst.rstdt",
|
19 |
+
"eng.sdrt.stac",
|
20 |
+
"deu.rst.pcc",
|
21 |
+
"eus.rst.ert",
|
22 |
+
"fas.rst.prstc",
|
23 |
+
"fra.sdrt.annodis",
|
24 |
+
"ita.pdtb.luna",
|
25 |
+
"nld.rst.nldt",
|
26 |
+
"por.pdtb.crpc",
|
27 |
+
"por.pdtb.tedm",
|
28 |
+
"por.rst.cstn",
|
29 |
+
"rus.rst.rrt",
|
30 |
+
"spa.rst.rststb",
|
31 |
+
"spa.rst.sctb",
|
32 |
+
"tha.pdtb.tdtb",
|
33 |
+
"tur.pdtb.tdb",
|
34 |
+
"tur.pdtb.tedm",
|
35 |
+
"zho.dep.scidtb",
|
36 |
+
"zho.pdtb.cdtb",
|
37 |
+
"zho.rst.gcdt",
|
38 |
+
"zho.rst.sctb",
|
39 |
+
]
|
40 |
+
|
41 |
+
|
42 |
+
_URLs = {f'{task}-{split}':f"https://raw.githubusercontent.com/disrpt/sharedtask2023/main/data/{task}/{task}_{split}.rels" \
|
43 |
+
for task in files for split in 'train dev test'.split()}
|
44 |
+
_URLs = {k:v for k,v in _URLs.items() if requests.get(v).status_code!=404}
|
45 |
+
|
46 |
+
class Dataset(datasets.GeneratorBasedBuilder):
|
47 |
+
|
48 |
+
BUILDER_CONFIGS = [
|
49 |
+
Config(
|
50 |
+
name=n,
|
51 |
+
data_dir=n
|
52 |
+
) for n in files
|
53 |
+
]
|
54 |
+
|
55 |
+
def _split_generators(self, dl_manager: datasets.DownloadManager):
|
56 |
+
data_file = dl_manager.download(_URLs)
|
57 |
+
train_key = self.config.data_dir+'-train'
|
58 |
+
if train_key in data_file:
|
59 |
+
train=datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": data_file[train_key]})
|
60 |
+
else:
|
61 |
+
train=[]
|
62 |
+
return train+[
|
63 |
+
datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filepath": data_file[self.config.data_dir+'-dev']}),
|
64 |
+
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": data_file[self.config.data_dir+'-test']}),
|
65 |
+
|
66 |
+
]
|
67 |
+
|
68 |
+
|
69 |
+
def _info(self): return datasets.DatasetInfo()
|
70 |
+
|
71 |
+
def _generate_examples(self, filepath):
|
72 |
+
with open(filepath, encoding="utf-8") as f:
|
73 |
+
reader = csv.DictReader(f,delimiter='\t')
|
74 |
+
for id_, row in enumerate(reader):
|
75 |
+
if id_ == 0:
|
76 |
+
continue
|
77 |
+
yield id_, row
|