bobox commited on
Commit
895ad6f
·
verified ·
1 Parent(s): 6421266

Upload generated eval data for eli5 (with 'eval_size' of 320)

Browse files

EVAL DATASET GENERATED FROM ORIGINAL DATASET'S TRAINING SPLIT (with 'eval_size' of 320)
Hard negative mining performed AFTER train/eval splitting original pool size: 325475; so, this eval split may contain negatives from training split)

Source dataset: sentence-transformers/eli5
(dataset name: eli5).
Hard negatives mined with model: sentence-transformers/all-MiniLM-L6-v2.
Mining params: max_score=0.8, num_negatives=1, relative_margin=0.05, absolute_margin=0.03, range_max=100.

README.md CHANGED
@@ -12,8 +12,11 @@ dataset_info:
12
  - name: train
13
  num_bytes: 240047421
14
  num_examples: 240963
15
- download_size: 156184942
16
- dataset_size: 240047421
 
 
 
17
  - config_name: natural-questions-1HN
18
  features:
19
  - name: query
@@ -104,6 +107,8 @@ configs:
104
  data_files:
105
  - split: train
106
  path: eli5-1HN/train-*
 
 
107
  - config_name: natural-questions-1HN
108
  data_files:
109
  - split: train
 
12
  - name: train
13
  num_bytes: 240047421
14
  num_examples: 240963
15
+ - name: eval
16
+ num_bytes: 310127
17
+ num_examples: 320
18
+ download_size: 156392336
19
+ dataset_size: 240357548
20
  - config_name: natural-questions-1HN
21
  features:
22
  - name: query
 
107
  data_files:
108
  - split: train
109
  path: eli5-1HN/train-*
110
+ - split: eval
111
+ path: eli5-1HN/eval-*
112
  - config_name: natural-questions-1HN
113
  data_files:
114
  - split: train
eli5-1HN/eval-00000-of-00001.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0015af1f44c4c90424e02da4bee15980fdfb7438d4975b6489b39f1caa29775f
3
+ size 207394