albertvillanova HF staff commited on
Commit
d1fb0b2
1 Parent(s): d310843

Delete loading script

Browse files
Files changed (1) hide show
  1. chr_en.py +0 -202
chr_en.py DELETED
@@ -1,202 +0,0 @@
1
- # coding=utf-8
2
- # Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
3
- #
4
- # Licensed under the Apache License, Version 2.0 (the "License");
5
- # you may not use this file except in compliance with the License.
6
- # You may obtain a copy of the License at
7
- #
8
- # http://www.apache.org/licenses/LICENSE-2.0
9
- #
10
- # Unless required by applicable law or agreed to in writing, software
11
- # distributed under the License is distributed on an "AS IS" BASIS,
12
- # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
- # See the License for the specific language governing permissions and
14
- # limitations under the License.
15
- """ChrEn: Cherokee-English Machine Translation data"""
16
-
17
-
18
- import openpyxl # noqa: requires this pandas optional dependency for reading xlsx files
19
- import pandas as pd
20
-
21
- import datasets
22
-
23
-
24
- _CITATION = """\
25
- @inproceedings{zhang2020chren,
26
- title={ChrEn: Cherokee-English Machine Translation for Endangered Language Revitalization},
27
- author={Zhang, Shiyue and Frey, Benjamin and Bansal, Mohit},
28
- booktitle={EMNLP2020},
29
- year={2020}
30
- }
31
- """
32
-
33
- _DESCRIPTION = """\
34
- ChrEn is a Cherokee-English parallel dataset to facilitate machine translation research between Cherokee and English.
35
- ChrEn is extremely low-resource contains 14k sentence pairs in total, split in ways that facilitate both in-domain and out-of-domain evaluation.
36
- ChrEn also contains 5k Cherokee monolingual data to enable semi-supervised learning.
37
- """
38
-
39
- _HOMEPAGE = "https://github.com/ZhangShiyue/ChrEn"
40
-
41
- _LICENSE = ""
42
-
43
- _URLs = {
44
- "monolingual_raw": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/raw/monolingual_data.xlsx",
45
- "parallel_raw": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/raw/parallel_data.xlsx",
46
- "monolingual_chr": "https://raw.githubusercontent.com/ZhangShiyue/ChrEn/main/data/monolingual/chr",
47
- "monolingual_en5000": "https://raw.githubusercontent.com/ZhangShiyue/ChrEn/main/data/monolingual/en5000",
48
- "monolingual_en10000": "https://raw.githubusercontent.com/ZhangShiyue/ChrEn/main/data/monolingual/en10000",
49
- "monolingual_en20000": "https://raw.githubusercontent.com/ZhangShiyue/ChrEn/main/data/monolingual/en20000",
50
- "monolingual_en50000": "https://raw.githubusercontent.com/ZhangShiyue/ChrEn/main/data/monolingual/en50000",
51
- "monolingual_en100000": "https://raw.githubusercontent.com/ZhangShiyue/ChrEn/main/data/monolingual/en100000",
52
- "parallel_train.chr": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/parallel/train.chr",
53
- "parallel_train.en": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/parallel/train.en",
54
- "parallel_dev.chr": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/parallel/dev.chr",
55
- "parallel_dev.en": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/parallel/dev.en",
56
- "parallel_out_dev.chr": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/parallel/out_dev.chr",
57
- "parallel_out_dev.en": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/parallel/out_dev.en",
58
- "parallel_test.chr": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/parallel/test.chr",
59
- "parallel_test.en": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/parallel/test.en",
60
- "parallel_out_test.chr": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/parallel/out_test.chr",
61
- "parallel_out_test.en": "https://github.com/ZhangShiyue/ChrEn/raw/main/data/parallel/out_test.en",
62
- }
63
-
64
-
65
- # TODO: Name of the dataset usually match the script name with CamelCase instead of snake_case
66
- class ChrEn(datasets.GeneratorBasedBuilder):
67
- """ChrEn: Cherokee-English Machine Translation data."""
68
-
69
- VERSION = datasets.Version("1.0.0")
70
-
71
- BUILDER_CONFIGS = [
72
- datasets.BuilderConfig(name="monolingual_raw", version=VERSION, description="Monolingual data with metadata"),
73
- datasets.BuilderConfig(name="parallel_raw", version=VERSION, description="Parallel data with metadata"),
74
- datasets.BuilderConfig(name="monolingual", version=VERSION, description="Monolingual data text only"),
75
- datasets.BuilderConfig(
76
- name="parallel", version=VERSION, description="Parallel data text pairs only with default split"
77
- ),
78
- ]
79
-
80
- DEFAULT_CONFIG_NAME = (
81
- "parallel" # It's not mandatory to have a default configuration. Just use one if it make sense.
82
- )
83
-
84
- def _info(self):
85
- if (
86
- self.config.name == "monolingual_raw"
87
- ): # This is the name of the configuration selected in BUILDER_CONFIGS above
88
- features = datasets.Features(
89
- {
90
- "text_sentence": datasets.Value("string"),
91
- "text_title": datasets.Value("string"),
92
- "speaker": datasets.Value("string"),
93
- "date": datasets.Value("int32"),
94
- "type": datasets.Value("string"),
95
- "dialect": datasets.Value("string"),
96
- }
97
- )
98
- elif (
99
- self.config.name == "parallel_raw"
100
- ): # This is the name of the configuration selected in BUILDER_CONFIGS above
101
- features = datasets.Features(
102
- {
103
- "line_number": datasets.Value("string"), # doesn't always map to a number
104
- "sentence_pair": datasets.Translation(languages=["en", "chr"]),
105
- "text_title": datasets.Value("string"),
106
- "speaker": datasets.Value("string"),
107
- "date": datasets.Value("int32"),
108
- "type": datasets.Value("string"),
109
- "dialect": datasets.Value("string"),
110
- }
111
- )
112
- elif (
113
- self.config.name == "parallel"
114
- ): # This is an example to show how to have different features for "first_domain" and "second_domain"
115
- features = datasets.Features(
116
- {
117
- "sentence_pair": datasets.Translation(languages=["en", "chr"]),
118
- }
119
- )
120
- elif (
121
- self.config.name == "monolingual"
122
- ): # This is an example to show how to have different features for "first_domain" and "second_domain"
123
- features = datasets.Features(
124
- {
125
- "sentence": datasets.Value("string"),
126
- }
127
- )
128
- return datasets.DatasetInfo(
129
- description=_DESCRIPTION,
130
- features=features, # Here we define them above because they are different between the two configurations
131
- supervised_keys=None,
132
- homepage=_HOMEPAGE,
133
- license=_LICENSE,
134
- citation=_CITATION,
135
- )
136
-
137
- def _split_generators(self, dl_manager):
138
- """Returns SplitGenerators."""
139
- data_dir = dl_manager.download(_URLs)
140
- if self.config.name in [
141
- "monolingual_raw",
142
- "parallel_raw",
143
- ]: # This is the name of the configuration selected in BUILDER_CONFIGS above
144
- return [
145
- datasets.SplitGenerator(
146
- name="full",
147
- gen_kwargs={
148
- "filepaths": data_dir,
149
- "split": "full",
150
- },
151
- )
152
- ]
153
- elif self.config.name == "monolingual":
154
- return [
155
- datasets.SplitGenerator(
156
- name=spl,
157
- gen_kwargs={
158
- "filepaths": data_dir,
159
- "split": spl,
160
- },
161
- )
162
- for spl in ["chr", "en5000", "en10000", "en20000", "en50000", "en100000"]
163
- ]
164
- else:
165
- return [
166
- datasets.SplitGenerator(
167
- name=spl,
168
- gen_kwargs={
169
- "filepaths": data_dir,
170
- "split": spl,
171
- },
172
- )
173
- for spl in ["train", "dev", "out_dev", "test", "out_test"]
174
- ]
175
-
176
- def _generate_examples(self, filepaths, split):
177
- if self.config.name == "monolingual_raw":
178
- keys = ["text_sentence", "text_title", "speaker", "date", "type", "dialect"]
179
- with open(filepaths["monolingual_raw"], "rb") as f:
180
- monolingual = pd.read_excel(f, engine="openpyxl")
181
- for id_, row in enumerate(monolingual.itertuples()):
182
- yield id_, dict(zip(keys, row[1:]))
183
- elif self.config.name == "parallel_raw":
184
- keys = ["line_number", "en_sent", "chr_sent", "text_title", "speaker", "date", "type", "dialect"]
185
- with open(filepaths["parallel_raw"], "rb") as f:
186
- parallel = pd.read_excel(f, engine="openpyxl")
187
- for id_, row in enumerate(parallel.itertuples()):
188
- res = dict(zip(keys, row[1:]))
189
- res["sentence_pair"] = {"en": res["en_sent"], "chr": res["chr_sent"]}
190
- res["line_number"] = str(res["line_number"])
191
- del res["en_sent"]
192
- del res["chr_sent"]
193
- yield id_, res
194
- elif self.config.name == "monolingual":
195
- f = open(filepaths[f"monolingual_{split}"], encoding="utf-8")
196
- for id_, line in enumerate(f):
197
- yield id_, {"sentence": line.strip()}
198
- elif self.config.name == "parallel":
199
- fi = open(filepaths[f"parallel_{split}.en"], encoding="utf-8")
200
- fo = open(filepaths[f"parallel_{split}.chr"], encoding="utf-8")
201
- for id_, (line_en, line_chr) in enumerate(zip(fi, fo)):
202
- yield id_, {"sentence_pair": {"en": line_en.strip(), "chr": line_chr.strip()}}