Datasets:
conversion to jsonl, removal of python code
Browse files- README.md +47 -0
- stats.txt +21 -0
- train.json.gz +3 -0
- validation.json.gz +3 -0
README.md
CHANGED
@@ -26,6 +26,40 @@ task_ids:
|
|
26 |
- open-domain-qa
|
27 |
- extractive-qa
|
28 |
- document-retrieval
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
29 |
train-eval-index:
|
30 |
- col_mapping:
|
31 |
answers:
|
@@ -205,6 +239,19 @@ Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)
|
|
205 |
|
206 |
D. Hládek, J. Staš, J. Juhár and T. Koctúr, "Slovak Dataset for Multilingual Question Answering," in IEEE Access, vol. 11, pp. 32869-32881, 2023, doi: [10.1109/ACCESS.2023.3262308.](https://doi.org/10.1109/ACCESS.2023.3262308)
|
207 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
208 |
### Contributions
|
209 |
|
210 |
Thanks to [@github-username](https://github.com/<github-username>) for adding this dataset.
|
|
|
26 |
- open-domain-qa
|
27 |
- extractive-qa
|
28 |
- document-retrieval
|
29 |
+
dataset_info:
|
30 |
+
config_name: skquad
|
31 |
+
features:
|
32 |
+
- name: id
|
33 |
+
dtype: string
|
34 |
+
- name: title
|
35 |
+
dtype: string
|
36 |
+
- name: context
|
37 |
+
dtype: string
|
38 |
+
- name: question
|
39 |
+
dtype: string
|
40 |
+
- name: answers
|
41 |
+
sequence:
|
42 |
+
- name: text
|
43 |
+
dtype: string
|
44 |
+
- name: answer_start
|
45 |
+
dtype: int32
|
46 |
+
splits:
|
47 |
+
- name: train
|
48 |
+
num_bytes: 111842848
|
49 |
+
num_examples: 81582
|
50 |
+
- name: validation
|
51 |
+
num_bytes: 12992195
|
52 |
+
num_examples: 9583
|
53 |
+
|
54 |
+
configs:
|
55 |
+
- config_name: skquad
|
56 |
+
default: true
|
57 |
+
data_files:
|
58 |
+
- split: train
|
59 |
+
path: train.json.gz
|
60 |
+
- split: validation
|
61 |
+
path: validation.json.gz
|
62 |
+
|
63 |
train-eval-index:
|
64 |
- col_mapping:
|
65 |
answers:
|
|
|
239 |
|
240 |
D. Hládek, J. Staš, J. Juhár and T. Koctúr, "Slovak Dataset for Multilingual Question Answering," in IEEE Access, vol. 11, pp. 32869-32881, 2023, doi: [10.1109/ACCESS.2023.3262308.](https://doi.org/10.1109/ACCESS.2023.3262308)
|
241 |
|
242 |
+
```
|
243 |
+
@ARTICLE{10082887,
|
244 |
+
author={Hládek, Daniel and Staš, Ján and Juhár, Jozef and Koctúr, Tomáš},
|
245 |
+
journal={IEEE Access},
|
246 |
+
title={Slovak Dataset for Multilingual Question Answering},
|
247 |
+
year={2023},
|
248 |
+
volume={11},
|
249 |
+
pages={32869-32881},
|
250 |
+
doi={10.1109/ACCESS.2023.3262308}
|
251 |
+
}
|
252 |
+
```
|
253 |
+
|
254 |
+
|
255 |
### Contributions
|
256 |
|
257 |
Thanks to [@github-username](https://github.com/<github-username>) for adding this dataset.
|
stats.txt
ADDED
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
## Trainset
|
2 |
+
Documents: 8377
|
3 |
+
Paragraphs: 22062
|
4 |
+
Questions: 81582
|
5 |
+
Answers: 65839
|
6 |
+
Unanswerable questions: 15877
|
7 |
+
Plausible answers: 15877
|
8 |
+
## Testset
|
9 |
+
Documents: 940
|
10 |
+
Paragraphs: 2568
|
11 |
+
Questions: 9583
|
12 |
+
Answers: 7822
|
13 |
+
Unanswerable questions: 1784
|
14 |
+
Plausible answers: 1784
|
15 |
+
## Total
|
16 |
+
Documents: 9317
|
17 |
+
Paragraphs: 24630
|
18 |
+
Questions: 91165
|
19 |
+
Answers: 73661
|
20 |
+
Unanswerable questions: 17661
|
21 |
+
Plausible answers: 17661
|
train.json.gz
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4f93dcf619ca9e46f5021220834e96b0a5a62a83ef3eefb6ee2aa6df286566c7
|
3 |
+
size 11006573
|
validation.json.gz
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e6ce488a85dd7c51ed5cc2393ff7a5a2b1cde08925967735e56e78e0968d3b2a
|
3 |
+
size 1264558
|