Update README.md
Browse files
README.md
CHANGED
|
@@ -1,5 +1,20 @@
|
|
| 1 |
---
|
| 2 |
base_model: bert-base-multilingual-uncased
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
| 4 |
|
| 5 |
# Language Detection Model
|
|
@@ -11,7 +26,30 @@ pretrained on [multilingual texts](https://huggingface.co/bert-base-multilingual
|
|
| 11 |
|
| 12 |
The model has been fine-tuned based on [Language Detection](https://www.kaggle.com/datasets/basilb2s/language-detection)
|
| 13 |
dataset found on *Kaggle*. The entire process of the dataset analysis as well as a complete description of the training procedure
|
| 14 |
-
can be found in [one of my *Kaggle* notebooks](https://www.kaggle.com/code/sergeypolivin/language-recognition-using-bert)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
## References
|
| 17 |
|
|
|
|
| 1 |
---
|
| 2 |
base_model: bert-base-multilingual-uncased
|
| 3 |
+
model-index:
|
| 4 |
+
- name: lang-recogn-model
|
| 5 |
+
results:
|
| 6 |
+
- task:
|
| 7 |
+
type: text-classification
|
| 8 |
+
dataset:
|
| 9 |
+
name: ai2_arc
|
| 10 |
+
type: ai2_arc
|
| 11 |
+
metrics:
|
| 12 |
+
- name: accuracy
|
| 13 |
+
type: accuracy
|
| 14 |
+
value: 0.9836
|
| 15 |
+
source:
|
| 16 |
+
name: Language Detection Dataset
|
| 17 |
+
url: https://www.kaggle.com/datasets/basilb2s/language-detection
|
| 18 |
---
|
| 19 |
|
| 20 |
# Language Detection Model
|
|
|
|
| 26 |
|
| 27 |
The model has been fine-tuned based on [Language Detection](https://www.kaggle.com/datasets/basilb2s/language-detection)
|
| 28 |
dataset found on *Kaggle*. The entire process of the dataset analysis as well as a complete description of the training procedure
|
| 29 |
+
can be found in [one of my *Kaggle* notebooks](https://www.kaggle.com/code/sergeypolivin/language-recognition-using-bert)
|
| 30 |
+
which has been used for the purpose of a faster model training on *GPU*.
|
| 31 |
+
|
| 32 |
+
## Supported languages
|
| 33 |
+
|
| 34 |
+
The model has been fine-tuned to detect one of the following 17 languages:
|
| 35 |
+
|
| 36 |
+
- Arabic
|
| 37 |
+
- Danish
|
| 38 |
+
- Dutch
|
| 39 |
+
- English
|
| 40 |
+
- French
|
| 41 |
+
- German
|
| 42 |
+
- Greek
|
| 43 |
+
- Hindi
|
| 44 |
+
- Italian
|
| 45 |
+
- Kannada
|
| 46 |
+
- Malayalam
|
| 47 |
+
- Portugeese
|
| 48 |
+
- Russian
|
| 49 |
+
- Spanish
|
| 50 |
+
- Sweedish
|
| 51 |
+
- Tamil
|
| 52 |
+
- Turkish
|
| 53 |
|
| 54 |
## References
|
| 55 |
|