Instructions to use ahmedabdelali/bert-base-qarib with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ahmedabdelali/bert-base-qarib with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="ahmedabdelali/bert-base-qarib")# Load model directly from transformers import AutoTokenizer, AutoModelForMaskedLM tokenizer = AutoTokenizer.from_pretrained("ahmedabdelali/bert-base-qarib") model = AutoModelForMaskedLM.from_pretrained("ahmedabdelali/bert-base-qarib", device_map="auto") - Inference
- Notebooks
- Google Colab
- Kaggle
Ahmed Abdelali commited on
Commit ·
a29b190
1
Parent(s): 6999b57
update config
Browse files- README.md +1 -1
- config.json +3 -0
README.md
CHANGED
|
@@ -17,7 +17,7 @@ widget:
|
|
| 17 |
## About QARiB
|
| 18 |
<img src="./Qarib_logo.png" width="100" align="left"/>
|
| 19 |
QCRI Arabic and Dialectal BERT (QARiB) model, was trained on a collection of ~ 420 Million tweets and ~ 180 Million sentences of text.
|
| 20 |
-
For
|
| 21 |
[Arabic GigaWord](url), [Abulkhair Arabic Corpus]() and [OPUS](http://opus.nlpl.eu/).
|
| 22 |
|
| 23 |
QARiB: Is the Arabic name for "Boat".
|
|
|
|
| 17 |
## About QARiB
|
| 18 |
<img src="./Qarib_logo.png" width="100" align="left"/>
|
| 19 |
QCRI Arabic and Dialectal BERT (QARiB) model, was trained on a collection of ~ 420 Million tweets and ~ 180 Million sentences of text.
|
| 20 |
+
For the tweets, the data was collected using twitter API and using language filter. `lang:ar`. For the text data, it was a combination from
|
| 21 |
[Arabic GigaWord](url), [Abulkhair Arabic Corpus]() and [OPUS](http://opus.nlpl.eu/).
|
| 22 |
|
| 23 |
QARiB: Is the Arabic name for "Boat".
|
config.json
CHANGED
|
@@ -1,4 +1,7 @@
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
| 2 |
"attention_probs_dropout_prob": 0.1,
|
| 3 |
"directionality": "bidi",
|
| 4 |
"hidden_act": "gelu",
|
|
|
|
| 1 |
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertModel"
|
| 4 |
+
],
|
| 5 |
"attention_probs_dropout_prob": 0.1,
|
| 6 |
"directionality": "bidi",
|
| 7 |
"hidden_act": "gelu",
|