carlosdanielhernandezmena commited on
Commit
decaadd
·
verified ·
1 Parent(s): b4ba0cb

Adding the train split

Browse files
commonvoice_benchmark_catalan_accents.py CHANGED
@@ -29,6 +29,8 @@ _LICENSE = "CC-BY-4.0, See https://creativecommons.org/licenses/by/4.0/"
29
 
30
  _BASE_DATA_DIR = "corpus/"
31
 
 
 
32
  _METADATA_BALEARIC_FEM = os.path.join(_BASE_DATA_DIR,"files","balearic_female.tsv")
33
  _METADATA_BALEARIC_MALE = os.path.join(_BASE_DATA_DIR,"files","balearic_male.tsv")
34
 
@@ -104,6 +106,8 @@ class CommonVoiceBenchmarkCatalanAccents(datasets.GeneratorBasedBuilder):
104
 
105
  def _split_generators(self, dl_manager):
106
 
 
 
107
  metadata_balearic_fem=dl_manager.download_and_extract(_METADATA_BALEARIC_FEM)
108
  metadata_balearic_male=dl_manager.download_and_extract(_METADATA_BALEARIC_MALE)
109
 
@@ -122,6 +126,9 @@ class CommonVoiceBenchmarkCatalanAccents(datasets.GeneratorBasedBuilder):
122
  tars_repo=dl_manager.download_and_extract(_TARS_REPO)
123
 
124
  hash_tar_files=defaultdict(dict)
 
 
 
125
 
126
  with open(tars_repo,'r') as f:
127
  hash_tar_files['balearic_fem']=[path.replace('\n','') for path in f]
@@ -148,7 +155,8 @@ class CommonVoiceBenchmarkCatalanAccents(datasets.GeneratorBasedBuilder):
148
  with open(tars_repo,'r') as f:
149
  hash_tar_files['valencian_male']=[path.replace('\n','') for path in f]
150
 
151
- hash_meta_paths={"balearic_fem":metadata_balearic_fem,
 
152
  "balearic_male":metadata_balearic_male,
153
  "central_female":metadata_central_female,
154
  "central_male":metadata_central_male,
@@ -161,7 +169,7 @@ class CommonVoiceBenchmarkCatalanAccents(datasets.GeneratorBasedBuilder):
161
 
162
  audio_paths = dl_manager.download(hash_tar_files)
163
 
164
- splits=["balearic_fem","balearic_male","central_female","central_male","northern_female",
165
  "northern_male","northwestern_female","northwestern_male","valencian_female","valencian_male"]
166
  local_extracted_audio_paths = (
167
  dl_manager.extract(audio_paths) if not dl_manager.is_streaming else
@@ -171,6 +179,14 @@ class CommonVoiceBenchmarkCatalanAccents(datasets.GeneratorBasedBuilder):
171
  )
172
 
173
  return [
 
 
 
 
 
 
 
 
174
  datasets.SplitGenerator(
175
  name="balearic_fem",
176
  gen_kwargs={
 
29
 
30
  _BASE_DATA_DIR = "corpus/"
31
 
32
+ _METADATA_TRAIN = os.path.join(_BASE_DATA_DIR,"files","train.tsv")
33
+
34
  _METADATA_BALEARIC_FEM = os.path.join(_BASE_DATA_DIR,"files","balearic_female.tsv")
35
  _METADATA_BALEARIC_MALE = os.path.join(_BASE_DATA_DIR,"files","balearic_male.tsv")
36
 
 
106
 
107
  def _split_generators(self, dl_manager):
108
 
109
+ metadata_train=dl_manager.download_and_extract(_METADATA_TRAIN)
110
+
111
  metadata_balearic_fem=dl_manager.download_and_extract(_METADATA_BALEARIC_FEM)
112
  metadata_balearic_male=dl_manager.download_and_extract(_METADATA_BALEARIC_MALE)
113
 
 
126
  tars_repo=dl_manager.download_and_extract(_TARS_REPO)
127
 
128
  hash_tar_files=defaultdict(dict)
129
+
130
+ with open(tars_repo,'r') as f:
131
+ hash_tar_files['train']=[path.replace('\n','') for path in f]
132
 
133
  with open(tars_repo,'r') as f:
134
  hash_tar_files['balearic_fem']=[path.replace('\n','') for path in f]
 
155
  with open(tars_repo,'r') as f:
156
  hash_tar_files['valencian_male']=[path.replace('\n','') for path in f]
157
 
158
+ hash_meta_paths={"train":metadata_train,
159
+ "balearic_fem":metadata_balearic_fem,
160
  "balearic_male":metadata_balearic_male,
161
  "central_female":metadata_central_female,
162
  "central_male":metadata_central_male,
 
169
 
170
  audio_paths = dl_manager.download(hash_tar_files)
171
 
172
+ splits=["train","balearic_fem","balearic_male","central_female","central_male","northern_female",
173
  "northern_male","northwestern_female","northwestern_male","valencian_female","valencian_male"]
174
  local_extracted_audio_paths = (
175
  dl_manager.extract(audio_paths) if not dl_manager.is_streaming else
 
179
  )
180
 
181
  return [
182
+ datasets.SplitGenerator(
183
+ name="train",
184
+ gen_kwargs={
185
+ "audio_archives":[dl_manager.iter_archive(archive) for archive in audio_paths["train"]],
186
+ "local_extracted_archives_paths": local_extracted_audio_paths["train"],
187
+ "metadata_paths": hash_meta_paths["train"],
188
+ }
189
+ ),
190
  datasets.SplitGenerator(
191
  name="balearic_fem",
192
  gen_kwargs={