Datasets:

Modalities:
Text
Formats:
parquet
Languages:
English
Libraries:
Datasets
pandas
License:
albertvillanova HF staff commited on
Commit
4e1ae2f
1 Parent(s): 2b25378

Convert dataset sizes from base 2 to base 10 in the dataset card (#2)

Browse files

- Convert dataset sizes from base 2 to base 10 in the dataset card (84d253609f23326d1204cf699f3006f559591f29)

Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -94,9 +94,9 @@ dataset_info:
94
  - **Repository:** https://github.com/text-machine-lab/quail
95
  - **Paper:** [Getting Closer to AI Complete Question Answering: A Set of Prerequisite Real Tasks](https://doi.org/10.1609/aaai.v34i05.6398 )
96
  - **Point of Contact:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
97
- - **Size of downloaded dataset files:** 6.11 MB
98
- - **Size of the generated dataset:** 28.25 MB
99
- - **Total amount of disk used:** 34.36 MB
100
 
101
  ### Dataset Summary
102
 
@@ -116,9 +116,9 @@ QuAIL is a reading comprehension dataset. QuAIL contains 15K multi-choice quest
116
 
117
  #### quail
118
 
119
- - **Size of downloaded dataset files:** 6.11 MB
120
- - **Size of the generated dataset:** 28.25 MB
121
- - **Total amount of disk used:** 34.36 MB
122
 
123
  An example of 'train' looks as follows.
124
  ```
 
94
  - **Repository:** https://github.com/text-machine-lab/quail
95
  - **Paper:** [Getting Closer to AI Complete Question Answering: A Set of Prerequisite Real Tasks](https://doi.org/10.1609/aaai.v34i05.6398 )
96
  - **Point of Contact:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
97
+ - **Size of downloaded dataset files:** 6.41 MB
98
+ - **Size of the generated dataset:** 29.62 MB
99
+ - **Total amount of disk used:** 36.03 MB
100
 
101
  ### Dataset Summary
102
 
 
116
 
117
  #### quail
118
 
119
+ - **Size of downloaded dataset files:** 6.41 MB
120
+ - **Size of the generated dataset:** 29.62 MB
121
+ - **Total amount of disk used:** 36.03 MB
122
 
123
  An example of 'train' looks as follows.
124
  ```