go-inoue commited on
Commit
7f64b4b
1 Parent(s): 4f23bb4

Update model name

Browse files
Files changed (1) hide show
  1. README.md +14 -14
README.md CHANGED
@@ -15,18 +15,18 @@ We release pre-trained language models for Modern Standard Arabic (MSA), dialect
15
  We also provide additional models that are pre-trained on a scaled-down set of the MSA variant (half, quarter, eighth, and sixteenth).
16
  The details are described in the paper *"[The Interplay of Variant, Size, and Task Type in Arabic Pre-trained Language Models](https://arxiv.org/abs/2103.06678)."*
17
 
18
- This model card describes **CAMeLBERT-MSA-quarter** (`bert-base-camelbert-msa-quarter`), a model pre-trained on a quarter of the full MSA dataset.
19
 
20
  ||Model|Variant|Size|#Word|
21
  |-|-|:-:|-:|-:|
22
- ||`bert-base-camelbert-mix`|CA,DA,MSA|167GB|17.3B|
23
- ||`bert-base-camelbert-ca`|CA|6GB|847M|
24
- ||`bert-base-camelbert-da`|DA|54GB|5.8B|
25
- ||`bert-base-camelbert-msa`|MSA|107GB|12.6B|
26
- ||`bert-base-camelbert-msa-half`|MSA|53GB|6.3B|
27
- |✔|`bert-base-camelbert-msa-quarter`|MSA|27GB|3.1B|
28
- ||`bert-base-camelbert-msa-eighth`|MSA|14GB|1.6B|
29
- ||`bert-base-camelbert-msa-sixteenth`|MSA|6GB|746M|
30
 
31
  ## Intended uses
32
  You can use the released model for either masked language modeling or next sentence prediction.
@@ -37,7 +37,7 @@ We release our fine-tuninig code [here](https://github.com/CAMeL-Lab/CAMeLBERT).
37
  You can use this model directly with a pipeline for masked language modeling:
38
  ```python
39
  >>> from transformers import pipeline
40
- >>> unmasker = pipeline('fill-mask', model='CAMeL-Lab/bert-base-camelbert-msa-quarter')
41
  >>> unmasker("الهدف من الحياة هو [MASK] .")
42
  [{'sequence': '[CLS] الهدف من الحياة هو الحياة. [SEP]',
43
  'score': 0.17437894642353058,
@@ -66,8 +66,8 @@ You can use this model directly with a pipeline for masked language modeling:
66
  Here is how to use this model to get the features of a given text in PyTorch:
67
  ```python
68
  from transformers import AutoTokenizer, AutoModel
69
- tokenizer = AutoTokenizer.from_pretrained('CAMeL-Lab/bert-base-camelbert-msa-quarter')
70
- model = AutoModel.from_pretrained('CAMeL-Lab/bert-base-camelbert-msa-quarter')
71
  text = "مرحبا يا عالم."
72
  encoded_input = tokenizer(text, return_tensors='pt')
73
  output = model(**encoded_input)
@@ -76,8 +76,8 @@ output = model(**encoded_input)
76
  and in TensorFlow:
77
  ```python
78
  from transformers import AutoTokenizer, TFAutoModel
79
- tokenizer = AutoTokenizer.from_pretrained('CAMeL-Lab/bert-base-camelbert-msa-quarter')
80
- model = TFAutoModel.from_pretrained('CAMeL-Lab/bert-base-camelbert-msa-quarter')
81
  text = "مرحبا يا عالم."
82
  encoded_input = tokenizer(text, return_tensors='tf')
83
  output = model(encoded_input)
 
15
  We also provide additional models that are pre-trained on a scaled-down set of the MSA variant (half, quarter, eighth, and sixteenth).
16
  The details are described in the paper *"[The Interplay of Variant, Size, and Task Type in Arabic Pre-trained Language Models](https://arxiv.org/abs/2103.06678)."*
17
 
18
+ This model card describes **CAMeLBERT-MSA-quarter** (`bert-base-arabic-camelbert-msa-quarter`), a model pre-trained on a quarter of the full MSA dataset.
19
 
20
  ||Model|Variant|Size|#Word|
21
  |-|-|:-:|-:|-:|
22
+ ||`bert-base-arabic-camelbert-mix`|CA,DA,MSA|167GB|17.3B|
23
+ ||`bert-base-arabic-camelbert-ca`|CA|6GB|847M|
24
+ ||`bert-base-arabic-camelbert-da`|DA|54GB|5.8B|
25
+ ||`bert-base-arabic-camelbert-msa`|MSA|107GB|12.6B|
26
+ ||`bert-base-arabic-camelbert-msa-half`|MSA|53GB|6.3B|
27
+ |✔|`bert-base-arabic-camelbert-msa-quarter`|MSA|27GB|3.1B|
28
+ ||`bert-base-arabic-camelbert-msa-eighth`|MSA|14GB|1.6B|
29
+ ||`bert-base-arabic-camelbert-msa-sixteenth`|MSA|6GB|746M|
30
 
31
  ## Intended uses
32
  You can use the released model for either masked language modeling or next sentence prediction.
 
37
  You can use this model directly with a pipeline for masked language modeling:
38
  ```python
39
  >>> from transformers import pipeline
40
+ >>> unmasker = pipeline('fill-mask', model='CAMeL-Lab/bert-base-arabic-camelbert-msa-quarter')
41
  >>> unmasker("الهدف من الحياة هو [MASK] .")
42
  [{'sequence': '[CLS] الهدف من الحياة هو الحياة. [SEP]',
43
  'score': 0.17437894642353058,
 
66
  Here is how to use this model to get the features of a given text in PyTorch:
67
  ```python
68
  from transformers import AutoTokenizer, AutoModel
69
+ tokenizer = AutoTokenizer.from_pretrained('CAMeL-Lab/bert-base-arabic-camelbert-msa-quarter')
70
+ model = AutoModel.from_pretrained('CAMeL-Lab/bert-base-arabic-camelbert-msa-quarter')
71
  text = "مرحبا يا عالم."
72
  encoded_input = tokenizer(text, return_tensors='pt')
73
  output = model(**encoded_input)
 
76
  and in TensorFlow:
77
  ```python
78
  from transformers import AutoTokenizer, TFAutoModel
79
+ tokenizer = AutoTokenizer.from_pretrained('CAMeL-Lab/bert-base-arabic-camelbert-msa-quarter')
80
+ model = TFAutoModel.from_pretrained('CAMeL-Lab/bert-base-arabic-camelbert-msa-quarter')
81
  text = "مرحبا يا عالم."
82
  encoded_input = tokenizer(text, return_tensors='tf')
83
  output = model(encoded_input)