Datasets:
Tasks:
Text Classification
Formats:
parquet
Sub-tasks:
natural-language-inference
Languages:
Russian
Size:
1M - 10M
cointegrated
commited on
Commit
•
2faa680
1
Parent(s):
ba183ce
Update README.md
Browse files
README.md
CHANGED
@@ -45,4 +45,34 @@ dataset_info:
|
|
45 |
---
|
46 |
# Dataset Card for "nli-rus-translated-v2021"
|
47 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
48 |
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
|
|
45 |
---
|
46 |
# Dataset Card for "nli-rus-translated-v2021"
|
47 |
|
48 |
+
This dataset was introduced in the Habr post
|
49 |
+
["Нейросети для Natural Language Inference (NLI): логические умозаключения на русском языке"](https://habr.com/ru/articles/582620/).
|
50 |
+
|
51 |
+
It is composed from various English NLI datasets automatically translated into Russian with two different methods.
|
52 |
+
|
53 |
+
Here are the sizes of the source datasets included into different splits:
|
54 |
+
|
55 |
+
| source | train | dev | test |
|
56 |
+
|:------------|--------:|------:|-------:|
|
57 |
+
| add_one_rte | 4991 | 387 | 0 |
|
58 |
+
| anli_r1 | 16946 | 1000 | 1000 |
|
59 |
+
| anli_r2 | 45460 | 1000 | 1000 |
|
60 |
+
| anli_r3 | 100459 | 1200 | 1200 |
|
61 |
+
| copa | 800 | 200 | 0 |
|
62 |
+
| fever | 162330 | 20478 | 20343 |
|
63 |
+
| help | 29347 | 3355 | 3189 |
|
64 |
+
| iie | 281643 | 31232 | 0 |
|
65 |
+
| imppres | 10179 | 7661 | 7660 |
|
66 |
+
| joci | 8412 | 939 | 0 |
|
67 |
+
| mnli | 392662 | 19647 | 0 |
|
68 |
+
| monli | 2186 | 269 | 223 |
|
69 |
+
| mpe | 9000 | 1000 | 0 |
|
70 |
+
| qnli | 108436 | 5732 | 0 |
|
71 |
+
| scitail | 24900 | 2126 | 0 |
|
72 |
+
| sick | 9500 | 500 | 0 |
|
73 |
+
| snli | 549297 | 9831 | 0 |
|
74 |
+
|
75 |
+
Most of the original data were taken from the repository [felipessalvatore/NLI_datasets](https://github.com/felipessalvatore/NLI_datasets).
|
76 |
+
|
77 |
+
|
78 |
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|