Hugging Face's logo

Datasets:
szhuggingface
/
ag_news_test

Modalities:
Text
Formats:
parquet
Size:
100K - 1M
Tags:
ag_news
text-classification
stratified-sampling
low-resource
benchmarking
Libraries:
Datasets
pandas
License:
Dataset card Data Studio Files
xet
 Community
1
szhuggingface commited on
Commit
822fc45
·
verified ·
1 Parent(s): ff54802

Update README.md

Browse files

Updated Data Fields

Files changed (1) hide show
  1. README.md +6 -0
README.md CHANGED
@@ -36,6 +36,12 @@ The dataset contains six splits. A `test` split was created from 20% of the orig
36
  | `train_very_small` | 12,000 | 1/8 of the `train_full` set. |
37
  | `train_very_very_small` | 6,000 | 1/16 of the `train_full` set. |
38
 
 
 
 
 
 
 
39
  ### How to Use
40
 
41
  You can load the dataset using the `datasets` library with a single line of code:
 
36
  | `train_very_small` | 12,000 | 1/8 of the `train_full` set. |
37
  | `train_very_very_small` | 6,000 | 1/16 of the `train_full` set. |
38
 
39
+
40
+ #### Data Fields
41
+ - `text`: a `string` feature.
42
+ - `label`: a classification label, with possible values including `World` (0), `Sports` (1), `Business` (2), `Sci/Tech` (3).
43
+
44
+
45
  ### How to Use
46
 
47
  You can load the dataset using the `datasets` library with a single line of code: