Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# QARiB: QCRI Arabic and Dialectal BERT
|
| 2 |
|
| 3 |
## About QARiB
|
|
@@ -71,7 +85,6 @@ You can use this model directly with a pipeline for masked language modeling:
|
|
| 71 |
|**Named-Entity Recognition (NER) | 49.38% | 64.97% | 66.63% | 64.04% | 61.62% |
|
| 72 |
|**Offensive Language Detection | 83.14% | 88.07% | 88.97% | 88.19% | 91.94% |
|
| 73 |
|**Sentiment Analysis | 86.61% | 90.80% | 93.58% | 83.27% | 93.31% |
|
| 74 |
-
|---------------------------------------------------------------------------------|
|
| 75 |
|
| 76 |
## Model Weights and Vocab Download
|
| 77 |
|
|
@@ -91,3 +104,4 @@ Ahmed Abdelali, Sabit Hassan, Hamdy Mubarak, Kareem Darwish and Younes Samih
|
|
| 91 |
}
|
| 92 |
```
|
| 93 |
|
|
|
| 1 |
+
---
|
| 2 |
+
language: ar
|
| 3 |
+
tags:
|
| 4 |
+
- pytorch
|
| 5 |
+
- tf
|
| 6 |
+
datasets:
|
| 7 |
+
- arabic_billion_words
|
| 8 |
+
- open_subtitles
|
| 9 |
+
- twitter
|
| 10 |
+
metrics:
|
| 11 |
+
- f1
|
| 12 |
+
widget:
|
| 13 |
+
- text: " شو عندكم يا [MASK] ."
|
| 14 |
+
---
|
| 15 |
# QARiB: QCRI Arabic and Dialectal BERT
|
| 16 |
|
| 17 |
## About QARiB
|
| 85 |
|**Named-Entity Recognition (NER) | 49.38% | 64.97% | 66.63% | 64.04% | 61.62% |
|
| 86 |
|**Offensive Language Detection | 83.14% | 88.07% | 88.97% | 88.19% | 91.94% |
|
| 87 |
|**Sentiment Analysis | 86.61% | 90.80% | 93.58% | 83.27% | 93.31% |
|
|
|
|
| 88 |
|
| 89 |
## Model Weights and Vocab Download
|
| 90 |
|
| 104 |
}
|
| 105 |
```
|
| 106 |
|
| 107 |
+
|