Datasets:

tiedeman commited on
Commit
b732aa6
1 Parent(s): 537e384

new json file

Browse files
Files changed (2) hide show
  1. dataset_infos.json +0 -0
  2. tatoeba_mt.py +10 -10
dataset_infos.json CHANGED
The diff for this file is too large to render. See raw diff
 
tatoeba_mt.py CHANGED
@@ -112,7 +112,7 @@ class tatoebaMT(datasets.GeneratorBasedBuilder):
112
 
113
  files = {
114
  "test": os.path.join("test", f"tatoeba-test.{lang_pair}.tsv"),
115
- "validation": os.path.join("dev", f"tatoeba-dev.{lang_pair}.tsv")
116
  }
117
  data_dir = dl_manager.download_and_extract(files)
118
 
@@ -127,15 +127,15 @@ class tatoebaMT(datasets.GeneratorBasedBuilder):
127
  )
128
  output.append(test)
129
 
130
- # if os.path.exists(data_dir["validation"]):
131
- valid = datasets.SplitGenerator(
132
- name=datasets.Split.VALIDATION,
133
- gen_kwargs={
134
- "filepath": data_dir["validation"],
135
- "split": "validation"
136
- }
137
- )
138
- output.append(valid)
139
 
140
  return output
141
 
 
112
 
113
  files = {
114
  "test": os.path.join("test", f"tatoeba-test.{lang_pair}.tsv"),
115
+ # "validation": os.path.join("dev", f"tatoeba-dev.{lang_pair}.tsv")
116
  }
117
  data_dir = dl_manager.download_and_extract(files)
118
 
 
127
  )
128
  output.append(test)
129
 
130
+ if os.path.exists(data_dir["validation"]):
131
+ valid = datasets.SplitGenerator(
132
+ name=datasets.Split.VALIDATION,
133
+ gen_kwargs={
134
+ "filepath": data_dir["validation"],
135
+ "split": "validation"
136
+ }
137
+ )
138
+ output.append(valid)
139
 
140
  return output
141