Datasets:

Languages:
English
Multilinguality:
multilingual
Size Categories:
10M<n<100M
Language Creators:
crowdsourced
Annotations Creators:
no-annotation
Source Datasets:
original
ArXiv:
Tags:
text-search
License:
mariosasko commited on
Commit
0cb5550
1 Parent(s): 0c1b1a7

Update wiki_dpr.py

Browse files
Files changed (1) hide show
  1. wiki_dpr.py +2 -19
wiki_dpr.py CHANGED
@@ -32,11 +32,6 @@ The wikipedia articles were split into multiple, disjoint text blocks of 100 wor
32
 
33
  _LICENSE = """DPR is CC-BY-NC 4.0 licensed."""
34
 
35
- _DATA_URL = "https://dl.fbaipublicfiles.com/dpr/wikipedia_split/psgs_w100.tsv.gz"
36
-
37
- _NQ_VECTORS_URL = "https://dl.fbaipublicfiles.com/dpr/data/wiki_encoded/single/nq/wiki_passages_{i}"
38
-
39
- _MULTISET_VECTORS_URL = "https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_{i}"
40
 
41
  _DATA_TO_NUM_SHARDS = {
42
  "nq": 157,
@@ -135,6 +130,7 @@ class WikiDpr(datasets.ArrowBasedBuilder):
135
  supervised_keys=None,
136
  homepage="https://github.com/facebookresearch/DPR",
137
  citation=_CITATION,
 
138
  )
139
 
140
  def _split_generators(self, dl_manager):
@@ -166,21 +162,8 @@ class WikiDpr(datasets.ArrowBasedBuilder):
166
  def _download_post_processing_resources(self, split, resource_name, dl_manager):
167
  if resource_name == "embeddings_index":
168
  try:
169
- embeddings_name = self.config.embeddings_name
170
- if self.config.dummy:
171
- embeddings_name = "dummy." + embeddings_name
172
- index_file = os.path.join(
173
- "data",
174
- self.config.wiki_split,
175
- self.config.embeddings_name,
176
- (
177
- "HNSW128_SQ8-IP-{split}.faiss"
178
- if self.config.index_name == "exaxt"
179
- else "IVF4096_HNSW128_PQ128-IP-{split}.faiss"
180
- ).format(split=split)
181
- )
182
  downloaded_resources = dl_manager.download_and_extract(
183
- {"embeddings_index": index_file}
184
  )
185
  return downloaded_resources["embeddings_index"]
186
  except (FileNotFoundError, ConnectionError): # index doesn't exist
 
32
 
33
  _LICENSE = """DPR is CC-BY-NC 4.0 licensed."""
34
 
 
 
 
 
 
35
 
36
  _DATA_TO_NUM_SHARDS = {
37
  "nq": 157,
 
130
  supervised_keys=None,
131
  homepage="https://github.com/facebookresearch/DPR",
132
  citation=_CITATION,
133
+ license=_LICENSE,
134
  )
135
 
136
  def _split_generators(self, dl_manager):
 
162
  def _download_post_processing_resources(self, split, resource_name, dl_manager):
163
  if resource_name == "embeddings_index":
164
  try:
 
 
 
 
 
 
 
 
 
 
 
 
 
165
  downloaded_resources = dl_manager.download_and_extract(
166
+ {"embeddings_index": os.path.join("index", self.config.index_file.format(split=split))}
167
  )
168
  return downloaded_resources["embeddings_index"]
169
  except (FileNotFoundError, ConnectionError): # index doesn't exist