Update README.md
Browse files
README.md
CHANGED
@@ -6,36 +6,26 @@ widget:
|
|
6 |
- text: " السلام عليكم ورحمة[MASK] وبركاتة"
|
7 |
- text: " اهلا وسهلا بكم في [MASK] من سيربح المليون "
|
8 |
---
|
9 |
-
|
10 |
# Arabic BERT Model
|
11 |
-
|
12 |
**AraBERTMo** is an Arabic pre-trained language model based on [Google's BERT architechture](https://github.com/google-research/bert).
|
13 |
AraBERTMo_base uses the same BERT-Base config.
|
14 |
AraBERTMo_base now comes in 10 new variants
|
15 |
All models are available on the `HuggingFace` model page under the [Ebtihal](https://huggingface.co/Ebtihal/) name.
|
16 |
Checkpoints are available in PyTorch formats.
|
17 |
|
18 |
-
|
19 |
## Pretraining Corpus
|
20 |
-
|
21 |
`AraBertMo_base_V1' model was pre-trained on ~3 million words:
|
22 |
-
|
23 |
- [OSCAR](https://traces1.inria.fr/oscar/) - Arabic version "unshuffled_deduplicated_ar".
|
24 |
|
25 |
-
|
26 |
## Training results
|
27 |
-
|
28 |
this model achieves the following results:
|
29 |
|
30 |
| Task | Num examples | Num Epochs | Batch Size | steps | Wall time | training loss|
|
31 |
|:----:|:----:|:----:|:----:|:-----:|:----:|:-----:|
|
32 |
| Fill-Mask| 10010| 1 | 64 | 157 | 2m 2s | 9.0183 |
|
33 |
|
34 |
-
|
35 |
## Load Pretrained Model
|
36 |
-
|
37 |
You can use this model by installing `torch` or `tensorflow` and Huggingface library `transformers`. And you can use it directly by initializing it like this:
|
38 |
-
|
39 |
```python
|
40 |
from transformers import AutoTokenizer, AutoModel
|
41 |
tokenizer = AutoTokenizer.from_pretrained("Ebtihal/AraBertMo_base_V1")
|
|
|
6 |
- text: " السلام عليكم ورحمة[MASK] وبركاتة"
|
7 |
- text: " اهلا وسهلا بكم في [MASK] من سيربح المليون "
|
8 |
---
|
|
|
9 |
# Arabic BERT Model
|
|
|
10 |
**AraBERTMo** is an Arabic pre-trained language model based on [Google's BERT architechture](https://github.com/google-research/bert).
|
11 |
AraBERTMo_base uses the same BERT-Base config.
|
12 |
AraBERTMo_base now comes in 10 new variants
|
13 |
All models are available on the `HuggingFace` model page under the [Ebtihal](https://huggingface.co/Ebtihal/) name.
|
14 |
Checkpoints are available in PyTorch formats.
|
15 |
|
|
|
16 |
## Pretraining Corpus
|
|
|
17 |
`AraBertMo_base_V1' model was pre-trained on ~3 million words:
|
|
|
18 |
- [OSCAR](https://traces1.inria.fr/oscar/) - Arabic version "unshuffled_deduplicated_ar".
|
19 |
|
|
|
20 |
## Training results
|
|
|
21 |
this model achieves the following results:
|
22 |
|
23 |
| Task | Num examples | Num Epochs | Batch Size | steps | Wall time | training loss|
|
24 |
|:----:|:----:|:----:|:----:|:-----:|:----:|:-----:|
|
25 |
| Fill-Mask| 10010| 1 | 64 | 157 | 2m 2s | 9.0183 |
|
26 |
|
|
|
27 |
## Load Pretrained Model
|
|
|
28 |
You can use this model by installing `torch` or `tensorflow` and Huggingface library `transformers`. And you can use it directly by initializing it like this:
|
|
|
29 |
```python
|
30 |
from transformers import AutoTokenizer, AutoModel
|
31 |
tokenizer = AutoTokenizer.from_pretrained("Ebtihal/AraBertMo_base_V1")
|