Update README.md
Browse files
README.md
CHANGED
|
@@ -12,7 +12,7 @@ pipeline_tag: token-classification
|
|
| 12 |
|
| 13 |
- tokenizer for hindi language
|
| 14 |
|
| 15 |
-
####
|
| 16 |
|
| 17 |
```py
|
| 18 |
from transformers import AutoTokenizer
|
|
@@ -28,18 +28,21 @@ encoded_str = hi_tokenizer.encode(hi_str)
|
|
| 28 |
decoded_str = hi_tokenizer.decode(encoded_str)
|
| 29 |
```
|
| 30 |
|
| 31 |
-
####
|
| 32 |
|
| 33 |
- hi
|
| 34 |
|
| 35 |
-
####
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
|
| 37 |
- trained on BHAAV (hi sentiment analysis dataset)
|
| 38 |
- dataset source: [Bhaav](https://github.com/midas-research/bhaav)
|
| 39 |
- Hindi text corpus (20,304 sentences)
|
| 40 |
|
| 41 |
-
|
| 42 |
-
#### citation
|
| 43 |
|
| 44 |
```shell
|
| 45 |
@article{kumar2019bhaav,
|
|
|
|
| 12 |
|
| 13 |
- tokenizer for hindi language
|
| 14 |
|
| 15 |
+
#### Usage
|
| 16 |
|
| 17 |
```py
|
| 18 |
from transformers import AutoTokenizer
|
|
|
|
| 28 |
decoded_str = hi_tokenizer.decode(encoded_str)
|
| 29 |
```
|
| 30 |
|
| 31 |
+
#### Language
|
| 32 |
|
| 33 |
- hi
|
| 34 |
|
| 35 |
+
#### Training
|
| 36 |
+
|
| 37 |
+
- For training see [Train BertWordPieceTokenizer](https://gist.github.com/kjdeveloper8/57d9e16848cd77df778804c9e2214a78)
|
| 38 |
+
|
| 39 |
+
#### Dataset
|
| 40 |
|
| 41 |
- trained on BHAAV (hi sentiment analysis dataset)
|
| 42 |
- dataset source: [Bhaav](https://github.com/midas-research/bhaav)
|
| 43 |
- Hindi text corpus (20,304 sentences)
|
| 44 |
|
| 45 |
+
#### Citation
|
|
|
|
| 46 |
|
| 47 |
```shell
|
| 48 |
@article{kumar2019bhaav,
|