Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,37 @@
|
|
1 |
---
|
2 |
-
license:
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
license: unknown
|
3 |
+
language:
|
4 |
+
- 'no'
|
5 |
---
|
6 |
+
|
7 |
+
## Dataset Card for NO-ConvAI2
|
8 |
+
|
9 |
+
NO-ConvAI2 is an open-domain human-to-bot conversational dataset machine translated from [ConvAI2](https://parl.ai/projects/convai2/).
|
10 |
+
|
11 |
+
In this dataset, the dialog_ids and turn_ids are omitted. Each line in the text are written into *Bot | Human* format.
|
12 |
+
|
13 |
+
## Data Split
|
14 |
+
The dataset is split into train and test sets.
|
15 |
+
| | #conversation_pairs |
|
16 |
+
|-------|---------------------|
|
17 |
+
| train | 253937 |
|
18 |
+
| test | 28658 |
|
19 |
+
|
20 |
+
|
21 |
+
## Licensing Information
|
22 |
+
|
23 |
+
This dataset is built upon the existing datasets. We therefore follow its original license information.
|
24 |
+
|
25 |
+
## Citation Information
|
26 |
+
|
27 |
+
Please cite original ConvAI2 dataset:
|
28 |
+
```
|
29 |
+
@inproceedings{dinan2020second,
|
30 |
+
title={The second conversational intelligence challenge (convai2)},
|
31 |
+
author={Dinan, Emily and Logacheva, Varvara and Malykh, Valentin and Miller, Alexander and Shuster, Kurt and Urbanek, Jack and Kiela, Douwe and Szlam, Arthur and Serban, Iulian and Lowe, Ryan and others},
|
32 |
+
booktitle={The NeurIPS'18 Competition: From Machine Learning to Intelligent Conversations},
|
33 |
+
pages={187--208},
|
34 |
+
year={2020},
|
35 |
+
organization={Springer}
|
36 |
+
}
|
37 |
+
```
|