Datasets:

License:
Mofe commited on
Commit
0773459
1 Parent(s): 5d8cb6a

Update config

Browse files
Files changed (2) hide show
  1. README.md +1 -1
  2. ciral.py +11 -2
README.md CHANGED
@@ -39,7 +39,7 @@ This dataset repo contains only the queries and relevance judgements. The corpus
39
  ```
40
  ciral_dataset = load_dataset("ciral/ciral", "hausa") #or swahili, somali, yoruba
41
 
42
- for data in ciral_data['dev']: # or 'testA' or 'testB'
43
  query_id = data['query_id']
44
  query = data['query']
45
  pos_qrels = data['positive_passages']
 
39
  ```
40
  ciral_dataset = load_dataset("ciral/ciral", "hausa") #or swahili, somali, yoruba
41
 
42
+ for data in ciral_dataset['dev']: # or 'testA' or 'testB'
43
  query_id = data['query_id']
44
  query = data['query']
45
  pos_qrels = data['positive_passages']
ciral.py CHANGED
@@ -108,6 +108,15 @@ class CIRAL(datasets.GeneratorBasedBuilder):
108
  'text': datasets.Value("string"),
109
  }],
110
  "negative_passages": [{
 
 
 
 
 
 
 
 
 
111
  "docid": datasets.Value("string"),
112
  "text": datasets.Value("string")}]
113
  }
@@ -144,7 +153,7 @@ class CIRAL(datasets.GeneratorBasedBuilder):
144
  datasets.SplitGenerator(
145
  name='testB',
146
  gen_kwargs={
147
- 'filepaths': downloaded_files['testA'],
148
  },
149
  ),
150
  ]
@@ -152,7 +161,7 @@ class CIRAL(datasets.GeneratorBasedBuilder):
152
  # method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
153
  def _generate_examples(self, filepaths):
154
  lang = self.config.name
155
- corpus = datasets.load_dataset('ciral/ciral-corpus', lang)
156
  docid2doc = {doc['docid']: doc['text'] for doc in corpus}
157
 
158
  query_file, qrel_file, pools_file = (filepaths) if len(filepaths) == 3 else (filepaths[0], filepaths[1], None)
 
108
  'text': datasets.Value("string"),
109
  }],
110
  "negative_passages": [{
111
+ "docid": datasets.Value("string"),
112
+ "text": datasets.Value("string")
113
+ }],
114
+
115
+ "pools_positive_passages": [{
116
+ 'docid': datasets.Value("string"),
117
+ 'text': datasets.Value("string"),
118
+ }],
119
+ "pools_negative_passages": [{
120
  "docid": datasets.Value("string"),
121
  "text": datasets.Value("string")}]
122
  }
 
153
  datasets.SplitGenerator(
154
  name='testB',
155
  gen_kwargs={
156
+ 'filepaths': downloaded_files['testB'],
157
  },
158
  ),
159
  ]
 
161
  # method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
162
  def _generate_examples(self, filepaths):
163
  lang = self.config.name
164
+ corpus = datasets.load_dataset('ciral/ciral-corpus', lang)['train']
165
  docid2doc = {doc['docid']: doc['text'] for doc in corpus}
166
 
167
  query_file, qrel_file, pools_file = (filepaths) if len(filepaths) == 3 else (filepaths[0], filepaths[1], None)