Hugging Face's logo

Datasets:
allenai
/
olmo-mix-1124

Tasks:
Text Generation
Modalities:
Text
Languages:
English
Size:
100M - 1B
License:
Dataset card Data Studio Files
xet
Community
15
Payoto commited on
Commit
c65def7
·
verified ·
1 Parent(s): 8162bd7

Correctly specify columns for dclm and default

Browse files

The columns of DCLM are a superset of the columns of the other samples so need to be made the default.

After this change the dataset can be loaded with all it's splits - test script below:

```python
olmo_stream_dict = {}
for split in ["default", "arxiv", "dclm", "starcoder", "wiki", "pes2o", "open-web-math", "algebraic-stack"]:
print(f"Loading split {split}...")
if split not in olmo_stream_dict:
olmo_stream_dict[split] = load_dataset(
"allenai/olmo-mix-1124",
split="train",
name=split,
verification_mode="no_checks",
streaming=True,
)
print(f"Loaded {olmo_stream_dict[split].num_shards} items in split {split}")

try:
for item in olmo_stream_dict[split]:
print(list(item.keys()))
break
except Exception as e:
print(f"Error iterating over split {split}: {e}")
try:
for item in olmo_stream_dict[split].shuffle(seed=42):
# This usually lands in dcml in default split - would error out
print("succesfully iterated shuffled split")
break
except Exception as e:
print(f"Error iterating over shuffled split {split}: {e}")
```

Files changed (1) hide show
  1. README.md +41 -17
README.md CHANGED
@@ -14,28 +14,40 @@ configs:
14
  - split: train
15
  path: data/**/*
16
  features:
 
 
 
 
 
 
 
 
17
  - name: text
18
  dtype: string
19
- - name: added
20
  dtype: string
21
- - name: created
22
  dtype: string
23
- - name: attributes
 
 
24
  dtype: string
25
  - name: doc
26
  dtype: string
27
- - name: id
28
  dtype: string
29
- - name: metadata
30
  dtype: string
31
  - name: source
32
  dtype: string
 
 
33
  - name: version
34
  dtype: string
35
  - config_name: algebraic-stack
36
  data_files:
37
  - split: train
38
- path: data/algebraic-stack/**/*
39
  features:
40
  - name: text
41
  dtype: string
@@ -54,11 +66,11 @@ configs:
54
  - name: source
55
  dtype: string
56
  - name: version
57
- dtype: string
58
  - config_name: arxiv
59
  data_files:
60
  - split: train
61
- path: data/arxiv/**/*
62
  features:
63
  - name: text
64
  dtype: string
@@ -81,30 +93,42 @@ configs:
81
  - config_name: dclm
82
  data_files:
83
  - split: train
84
- path: data/dclm/**/*
85
  features:
 
 
 
 
 
 
 
 
86
  - name: text
87
  dtype: string
88
- - name: added
89
  dtype: string
90
- - name: created
91
  dtype: string
92
- - name: attributes
 
 
93
  dtype: string
94
  - name: doc
95
  dtype: string
96
- - name: id
97
  dtype: string
98
- - name: metadata
99
  dtype: string
100
  - name: source
101
  dtype: string
 
 
102
  - name: version
103
  dtype: string
104
  - config_name: open-web-math
105
  data_files:
106
  - split: train
107
- path: data/open-web-math/**/*
108
  features:
109
  - name: text
110
  dtype: string
@@ -123,11 +147,11 @@ configs:
123
  - name: source
124
  dtype: string
125
  - name: version
126
- dtype: string
127
  - config_name: pes2o
128
  data_files:
129
  - split: train
130
- path: data/pes2o/*
131
  features:
132
  - name: text
133
  dtype: string
 
14
  - split: train
15
  path: data/**/*
16
  features:
17
+ - name: bff_contained_ngram_count_before_dedupe
18
+ dtype: int64
19
+ - name: language_id_whole_page_fasttext
20
+ dtype: string
21
+ - name: metadata
22
+ dtype: string
23
+ - name: previous_word_count
24
+ dtype: string
25
  - name: text
26
  dtype: string
27
+ - name: url
28
  dtype: string
29
+ - name: warcinfo
30
  dtype: string
31
+ - name: fasttext_openhermes_reddit_eli5_vs_rw_v2_bigram_200k_train_prob
32
+ dtype: string
33
+ - name: id
34
  dtype: string
35
  - name: doc
36
  dtype: string
37
+ - name: added
38
  dtype: string
39
+ - name: created
40
  dtype: string
41
  - name: source
42
  dtype: string
43
+ - name: attributes
44
+ dtype: string
45
  - name: version
46
  dtype: string
47
  - config_name: algebraic-stack
48
  data_files:
49
  - split: train
50
+ path: data/algebraic-stack/**/*
51
  features:
52
  - name: text
53
  dtype: string
 
66
  - name: source
67
  dtype: string
68
  - name: version
69
+ dtype: string
70
  - config_name: arxiv
71
  data_files:
72
  - split: train
73
+ path: data/arxiv/**/*
74
  features:
75
  - name: text
76
  dtype: string
 
93
  - config_name: dclm
94
  data_files:
95
  - split: train
96
+ path: data2/dclm/**/*
97
  features:
98
+ - name: bff_contained_ngram_count_before_dedupe
99
+ dtype: int64
100
+ - name: language_id_whole_page_fasttext
101
+ dtype: string
102
+ - name: metadata
103
+ dtype: string
104
+ - name: previous_word_count
105
+ dtype: string
106
  - name: text
107
  dtype: string
108
+ - name: url
109
  dtype: string
110
+ - name: warcinfo
111
  dtype: string
112
+ - name: fasttext_openhermes_reddit_eli5_vs_rw_v2_bigram_200k_train_prob
113
+ dtype: string
114
+ - name: id
115
  dtype: string
116
  - name: doc
117
  dtype: string
118
+ - name: added
119
  dtype: string
120
+ - name: created
121
  dtype: string
122
  - name: source
123
  dtype: string
124
+ - name: attributes
125
+ dtype: string
126
  - name: version
127
  dtype: string
128
  - config_name: open-web-math
129
  data_files:
130
  - split: train
131
+ path: data/open-web-math/**/*
132
  features:
133
  - name: text
134
  dtype: string
 
147
  - name: source
148
  dtype: string
149
  - name: version
150
+ dtype: string
151
  - config_name: pes2o
152
  data_files:
153
  - split: train
154
+ path: data/pes2o/*
155
  features:
156
  - name: text
157
  dtype: string