Hugging Face's logo

Datasets:
yandex
/
yambda

Modalities:
Tabular
Text
Formats:
parquet
Size:
1B - 10B
ArXiv:
Tags:
recsys
retrieval
dataset
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
22
tytskiy commited on
Commit
5ba1508
·
verified ·
1 Parent(s): fcb9470

Fix dataviewer

Browse files
Files changed (1) hide show
  1. README.md +56 -15
README.md CHANGED
@@ -9,22 +9,63 @@ size_categories:
9
  - 1B<n<10B
10
 
11
  configs:
12
- - config_name: default
13
  data_files:
14
- # - split: flat_5b
15
- # path: "flat/5b/likes.parquet"
16
- # - split: flat_500m
17
- # path: "flat/500m/*.parquet"
18
- - split: flat_50m
19
- path:
20
- - "flat/50m/likes.parquet"
21
- - "flat/50m/listens.parquet"
22
- # - split: sequential_5b
23
- # path: "sequential/5b/*.parquet"
24
- # - split: sequential_500m
25
- # path: "sequential/500m/*.parquet"
26
- # - split: sequential_50m
27
- # path: "sequential/50m/*.parquet"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
28
  ---
29
 
30
  # Yambda-5B — A Large-Scale Multi-modal Dataset for Ranking And Retrieval
 
9
  - 1B<n<10B
10
 
11
  configs:
12
+ - config_name: flat-50m
13
  data_files:
14
+ - split: likes
15
+ path: flat/50m/likes.parquet
16
+ - split: listens
17
+ path: flat/50m/listens.parquet
18
+ - split: unlikes
19
+ path: flat/50m/unlikes.parquet
20
+ - split: multi_event
21
+ path: flat/50m/multi_event.parquet
22
+ - split: dislikes
23
+ path: flat/50m/dislikes.parquet
24
+ - split: undislikes
25
+ path: flat/50m/undislikes.parquet
26
+ - config_name: flat-500m
27
+ data_files:
28
+ - split: likes
29
+ path: flat/500m/likes.parquet
30
+ - split: listens
31
+ path: flat/500m/listens.parquet
32
+ - split: unlikes
33
+ path: flat/500m/unlikes.parquet
34
+ - split: multi_event
35
+ path: flat/500m/multi_event.parquet
36
+ - split: dislikes
37
+ path: flat/500m/dislikes.parquet
38
+ - split: undislikes
39
+ path: flat/500m/undislikes.parquet
40
+ - config_name: sequential-50m
41
+ data_files:
42
+ - split: likes
43
+ path: sequential/50m/likes.parquet
44
+ - split: listens
45
+ path: sequential/50m/listens.parquet
46
+ - split: unlikes
47
+ path: sequential/50m/unlikes.parquet
48
+ - split: multi_event
49
+ path: sequential/50m/multi_event.parquet
50
+ - split: dislikes
51
+ path: sequential/50m/dislikes.parquet
52
+ - split: undislikes
53
+ path: sequential/50m/undislikes.parquet
54
+ - config_name: sequential-500m
55
+ data_files:
56
+ - split: likes
57
+ path: sequential/500m/likes.parquet
58
+ - split: listens
59
+ path: sequential/500m/listens.parquet
60
+ - split: unlikes
61
+ path: sequential/500m/unlikes.parquet
62
+ - split: multi_event
63
+ path: sequential/500m/multi_event.parquet
64
+ - split: dislikes
65
+ path: sequential/500m/dislikes.parquet
66
+ - split: undislikes
67
+ path: sequential/500m/undislikes.parquet
68
+
69
  ---
70
 
71
  # Yambda-5B — A Large-Scale Multi-modal Dataset for Ranking And Retrieval