holylovenia commited on
Commit
33a513b
1 Parent(s): 3e52fda

Upload parallel_id_nyo.py with huggingface_hub

Browse files
Files changed (1) hide show
  1. parallel_id_nyo.py +13 -17
parallel_id_nyo.py CHANGED
@@ -4,17 +4,17 @@ from typing import List
4
  import datasets
5
  import pandas as pd
6
 
7
- from nusacrowd.utils import schemas
8
- from nusacrowd.utils.configs import NusantaraConfig
9
- from nusacrowd.utils.constants import (
10
- DEFAULT_NUSANTARA_VIEW_NAME,
11
  DEFAULT_SOURCE_VIEW_NAME,
12
  Tasks,
13
  )
14
 
15
  _DATASETNAME = "parallel_id_nyo"
16
  _SOURCE_VIEW_NAME = DEFAULT_SOURCE_VIEW_NAME
17
- _UNIFIED_VIEW_NAME = DEFAULT_NUSANTARA_VIEW_NAME
18
 
19
  _LOCAL = False
20
  _LANGUAGES = ["ind", "abl"]
@@ -74,7 +74,7 @@ _URLs = {
74
  _SUPPORTED_TASKS = [Tasks.MACHINE_TRANSLATION]
75
 
76
  _SOURCE_VERSION = "1.0.0"
77
- _NUSANTARA_VERSION = "1.0.0"
78
 
79
  COL_INDONESIA = "indo"
80
  COL_LAMPUNG = "lampung"
@@ -84,18 +84,18 @@ class ParallelIdNyo(datasets.GeneratorBasedBuilder):
84
  """Dataset that contains Indonesian - Lampung language pairs."""
85
 
86
  BUILDER_CONFIGS = [
87
- NusantaraConfig(
88
  name="parallel_id_nyo_source",
89
  version=datasets.Version(_SOURCE_VERSION),
90
  description="Parallel Id-Nyo source schema",
91
  schema="source",
92
  subset_id="parallel_id_nyo",
93
  ),
94
- NusantaraConfig(
95
- name="parallel_id_nyo_nusantara_t2t",
96
- version=datasets.Version(_NUSANTARA_VERSION),
97
  description="Parallel Id-Nyo Nusantara schema",
98
- schema="nusantara_t2t",
99
  subset_id="parallel_id_nyo",
100
  ),
101
  ]
@@ -111,7 +111,7 @@ class ParallelIdNyo(datasets.GeneratorBasedBuilder):
111
  "label": datasets.Value("string"),
112
  }
113
  )
114
- elif self.config.schema == "nusantara_t2t":
115
  features = schemas.text2text_features
116
 
117
  return datasets.DatasetInfo(
@@ -146,7 +146,7 @@ class ParallelIdNyo(datasets.GeneratorBasedBuilder):
146
  "label": str(row[COL_LAMPUNG]).rstrip(),
147
  }
148
  yield idx, ex
149
- elif self.config.schema == "nusantara_t2t":
150
  for idx, row in df.iterrows():
151
  ex = {
152
  "id": str(idx),
@@ -158,7 +158,3 @@ class ParallelIdNyo(datasets.GeneratorBasedBuilder):
158
  yield idx, ex
159
  else:
160
  raise ValueError(f"Invalid config: {self.config.name}")
161
-
162
-
163
- if __name__ == "__main__":
164
- datasets.load_dataset(__file__)
 
4
  import datasets
5
  import pandas as pd
6
 
7
+ from seacrowd.utils import schemas
8
+ from seacrowd.utils.configs import SEACrowdConfig
9
+ from seacrowd.utils.constants import (
10
+ DEFAULT_SEACROWD_VIEW_NAME,
11
  DEFAULT_SOURCE_VIEW_NAME,
12
  Tasks,
13
  )
14
 
15
  _DATASETNAME = "parallel_id_nyo"
16
  _SOURCE_VIEW_NAME = DEFAULT_SOURCE_VIEW_NAME
17
+ _UNIFIED_VIEW_NAME = DEFAULT_SEACROWD_VIEW_NAME
18
 
19
  _LOCAL = False
20
  _LANGUAGES = ["ind", "abl"]
 
74
  _SUPPORTED_TASKS = [Tasks.MACHINE_TRANSLATION]
75
 
76
  _SOURCE_VERSION = "1.0.0"
77
+ _SEACROWD_VERSION = "2024.06.20"
78
 
79
  COL_INDONESIA = "indo"
80
  COL_LAMPUNG = "lampung"
 
84
  """Dataset that contains Indonesian - Lampung language pairs."""
85
 
86
  BUILDER_CONFIGS = [
87
+ SEACrowdConfig(
88
  name="parallel_id_nyo_source",
89
  version=datasets.Version(_SOURCE_VERSION),
90
  description="Parallel Id-Nyo source schema",
91
  schema="source",
92
  subset_id="parallel_id_nyo",
93
  ),
94
+ SEACrowdConfig(
95
+ name="parallel_id_nyo_seacrowd_t2t",
96
+ version=datasets.Version(_SEACROWD_VERSION),
97
  description="Parallel Id-Nyo Nusantara schema",
98
+ schema="seacrowd_t2t",
99
  subset_id="parallel_id_nyo",
100
  ),
101
  ]
 
111
  "label": datasets.Value("string"),
112
  }
113
  )
114
+ elif self.config.schema == "seacrowd_t2t":
115
  features = schemas.text2text_features
116
 
117
  return datasets.DatasetInfo(
 
146
  "label": str(row[COL_LAMPUNG]).rstrip(),
147
  }
148
  yield idx, ex
149
+ elif self.config.schema == "seacrowd_t2t":
150
  for idx, row in df.iterrows():
151
  ex = {
152
  "id": str(idx),
 
158
  yield idx, ex
159
  else:
160
  raise ValueError(f"Invalid config: {self.config.name}")