Fix typo

README.md

@@ -3,7 +3,7 @@ language:
 - ar
 license: apache-2.0
 widget:
- - text: "الهدف من الحياة هو [MASK] ."
+ - text: "\UTF{0627}\UTF{0644}\UTF{0647}\UTF{062F}\UTF{0641} \UTF{0645}\UTF{0646} \UTF{0627}\UTF{0644}\UTF{062D}\UTF{064A}\UTF{0627}\UTF{0629} \UTF{0647}\UTF{0648} [MASK] ."
 ---
 
 # bert-base-camelbert-msa
@@ -18,7 +18,7 @@ We release eight models with different sizes and variants as follows:
 |-|-|:-:|-:|-:|
 ||`bert-base-camelbert-mix`|CA,DA,MSA|167GB|17.3B|
 ||`bert-base-camelbert-ca`|CA|6GB|847M|
-|✔|`bert-base-camelbert-da`|DA|54GB|5.8B|
+|\UTF{2714}|`bert-base-camelbert-da`|DA|54GB|5.8B|
 ||`bert-base-camelbert-msa`|MSA|107GB|12.6B|
 ||`bert-base-camelbert-msa-half`|MSA|53GB|6.3B|
 ||`bert-base-camelbert-msa-quarter`|MSA|27GB|3.1B|
@@ -37,27 +37,27 @@ You can use this model directly with a pipeline for masked language modeling:
 ```python
 >>> from transformers import pipeline
 >>> unmasker = pipeline('fill-mask', model='CAMeL-Lab/bert-base-camelbert-da')
->>> unmasker("الهدف من الحياة هو [MASK] .")
-[{'sequence': '[CLS] الهدف من الحياة هو.. [SEP]',
+>>> unmasker("\UTF{0627}\UTF{0644}\UTF{0647}\UTF{062F}\UTF{0641} \UTF{0645}\UTF{0646} \UTF{0627}\UTF{0644}\UTF{062D}\UTF{064A}\UTF{0627}\UTF{0629} \UTF{0647}\UTF{0648} [MASK] .")
+[{'sequence': '[CLS] \UTF{0627}\UTF{0644}\UTF{0647}\UTF{062F}\UTF{0641} \UTF{0645}\UTF{0646} \UTF{0627}\UTF{0644}\UTF{062D}\UTF{064A}\UTF{0627}\UTF{0629} \UTF{0647}\UTF{0648}.. [SEP]',
   'score': 0.062508225440979,
   'token': 18,
   'token_str': '.'},
- {'sequence': '[CLS] الهدف من الحياة هو الموت. [SEP]',
+ {'sequence': '[CLS] \UTF{0627}\UTF{0644}\UTF{0647}\UTF{062F}\UTF{0641} \UTF{0645}\UTF{0646} \UTF{0627}\UTF{0644}\UTF{062D}\UTF{064A}\UTF{0627}\UTF{0629} \UTF{0647}\UTF{0648} \UTF{0627}\UTF{0644}\UTF{0645}\UTF{0648}\UTF{062A}. [SEP]',
   'score': 0.033172328025102615,
   'token': 4295,
-  'token_str': 'الموت'},
- {'sequence': '[CLS] الهدف من الحياة هو الحياة. [SEP]',
+  'token_str': '\UTF{0627}\UTF{0644}\UTF{0645}\UTF{0648}\UTF{062A}'},
+ {'sequence': '[CLS] \UTF{0627}\UTF{0644}\UTF{0647}\UTF{062F}\UTF{0641} \UTF{0645}\UTF{0646} \UTF{0627}\UTF{0644}\UTF{062D}\UTF{064A}\UTF{0627}\UTF{0629} \UTF{0647}\UTF{0648} \UTF{0627}\UTF{0644}\UTF{062D}\UTF{064A}\UTF{0627}\UTF{0629}. [SEP]',
   'score': 0.029575437307357788,
   'token': 3696,
-  'token_str': 'الحياة'},
- {'sequence': '[CLS] الهدف من الحياة هو الرحيل. [SEP]',
+  'token_str': '\UTF{0627}\UTF{0644}\UTF{062D}\UTF{064A}\UTF{0627}\UTF{0629}'},
+ {'sequence': '[CLS] \UTF{0627}\UTF{0644}\UTF{0647}\UTF{062F}\UTF{0641} \UTF{0645}\UTF{0646} \UTF{0627}\UTF{0644}\UTF{062D}\UTF{064A}\UTF{0627}\UTF{0629} \UTF{0647}\UTF{0648} \UTF{0627}\UTF{0644}\UTF{0631}\UTF{062D}\UTF{064A}\UTF{0644}. [SEP]',
   'score': 0.02724040113389492,
   'token': 11449,
-  'token_str': 'الرحيل'},
- {'sequence': '[CLS] الهدف من الحياة هو الحب. [SEP]',
+  'token_str': '\UTF{0627}\UTF{0644}\UTF{0631}\UTF{062D}\UTF{064A}\UTF{0644}'},
+ {'sequence': '[CLS] \UTF{0627}\UTF{0644}\UTF{0647}\UTF{062F}\UTF{0641} \UTF{0645}\UTF{0646} \UTF{0627}\UTF{0644}\UTF{062D}\UTF{064A}\UTF{0627}\UTF{0629} \UTF{0647}\UTF{0648} \UTF{0627}\UTF{0644}\UTF{062D}\UTF{0628}. [SEP]',
   'score': 0.01564178802073002,
   'token': 3088,
-  'token_str': 'الحب'}]
+  'token_str': '\UTF{0627}\UTF{0644}\UTF{062D}\UTF{0628}'}]
 ```
 
 Here is how to use this model to get the features of a given text in PyTorch:
@@ -65,7 +65,7 @@ Here is how to use this model to get the features of a given text in PyTorch:
 from transformers import AutoTokenizer, AutoModel
 tokenizer = AutoTokenizer.from_pretrained('CAMeL-Lab/bert-base-camelbert-da')
 model = AutoModel.from_pretrained('CAMeL-Lab/bert-base-camelbert-da')
-text = "مرحبا يا عالم."
+text = "\UTF{0645}\UTF{0631}\UTF{062D}\UTF{0628}\UTF{0627} \UTF{064A}\UTF{0627} \UTF{0639}\UTF{0627}\UTF{0644}\UTF{0645}."
 encoded_input = tokenizer(text, return_tensors='pt')
 output = model(**encoded_input)
 ```
@@ -75,18 +75,14 @@ and in TensorFlow:
 from transformers import AutoTokenizer, TFAutoModel
 tokenizer = AutoTokenizer.from_pretrained('CAMeL-Lab/bert-base-camelbert-da')
 model = TFAutoModel.from_pretrained('CAMeL-Lab/bert-base-camelbert-da')
-text = "مرحبا يا عالم."
+text = "\UTF{0645}\UTF{0631}\UTF{062D}\UTF{0628}\UTF{0627} \UTF{064A}\UTF{0627} \UTF{0639}\UTF{0627}\UTF{0644}\UTF{0645}."
 encoded_input = tokenizer(text, return_tensors='tf')
 output = model(encoded_input)
 ```
 
 ## Training data
-- MSA
-  - [The Arabic Gigaword Fifth Edition](https://catalog.ldc.upenn.edu/LDC2011T11)
-  - [Abu El-Khair Corpus](http://www.abuelkhair.net/index.php/en/arabic/abu-el-khair-corpus)
-  - [OSIAN corpus](https://vlo.clarin.eu/search;jsessionid=31066390B2C9E8C6304845BA79869AC1?1&q=osian)
-  - [Arabic Wikipedia](https://archive.org/details/arwiki-20190201)
-  - The unshuffled version of the Arabic [OSCAR corpus](https://oscar-corpus.com/)
+- DA
+  - A collection of dialectal Arabic data described in our paper.
 
 ## Training procedure
 We use [the original implementation](https://github.com/google-research/bert) released by Google for pre-training.