elmadany commited on
Commit
34455d5
1 Parent(s): 5c2860f

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +42 -0
README.md ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # AraT5-msa-small
2
+ <img src="https://raw.githubusercontent.com/UBC-NLP/araT5/main/AraT5_logo.jpg" alt="drawing" width="30%" height="30%" align="right"/>
3
+
4
+ **AraT5-base** is one of three models described in our [**AraT5: Text-to-Text Transformers for Arabic Language Understanding and Generation**
5
+ ](https://arxiv.org/abs/2109.12068). In this paper, we introduce three powerful Arabic-specific text-to-text transformer models trained on large Modern Standard Arabic (MSA) and/or Dialectal Arabic (DA) data. **AraT5** is trained on 248GB of text (29B tokens) of MSA and DA, **AraT5-msa** is trained on 70GB of text (7.1B tokens) from MSA data, and **AraT5-tweet** is trained on 178Gb of text (21.9B tokens) from 1.5B Arabic tweets which contains multiple varieties of dialectical Arabic.
6
+
7
+ In addition, we provide the three models on two architectures small and base. For all models, we use a learning rate of 0.01, a batch size of 128 sequences, and a maximum sequence length of 512 whereas AraT5-tweet 128 maximum sequence is used. Hence, the original implementation of T5 in the TensorFlow framework is used to train the models. We train the models for 1M steps.8 Training took ∼ 80 days on 1 on Google Cloud TPU with 8 cores (v3.8) from TensorFlow Research Cloud (TFRC).
8
+
9
+
10
+ # How to use AraT5 models
11
+ [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1GFOGolWPIfDvYdSNdGFrOXwu3Gu28k2b?usp=sharing)This is an example for fine-tuning **AraT5-base** for News Title Generation on the Aranews dataset
12
+
13
+ For more details, please visit our own [GitHub](https://github.com/UBC-NLP/araT5).
14
+
15
+ # AraT5 Models Checkpoints
16
+
17
+ AraT5 Pytorch and TensorFlow checkpoints are available on the Huggingface website for direct download and use ```exclusively for research```. ```For commercial use, please contact the authors via email @ (muhammad.mageed[at]ubc[dot]ca).```
18
+
19
+ | **Model** | **Link** |
20
+ |---------|:------------------:|
21
+ | **AraT5-base** | [https://huggingface.co/UBC-NLP/AraT5-base](https://huggingface.co/UBC-NLP/AraT5-base) |
22
+ | **AraT5-msa-base** | [https://huggingface.co/UBC-NLP/AraT5-msa-base](https://huggingface.co/UBC-NLP/AraT5-msa-base) |
23
+ | **AraT5-tweet-base** | [https://huggingface.co/UBC-NLP/AraT5-tweet-base](https://huggingface.co/UBC-NLP/AraT5-tweet-base) |
24
+ | **AraT5-msa-small** | [https://huggingface.co/UBC-NLP/AraT5-msa-small](https://huggingface.co/UBC-NLP/AraT5-msa-small) |
25
+ | **AraT5-tweet-small**| [https://huggingface.co/UBC-NLP/AraT5-tweet-small](https://huggingface.co/UBC-NLP/AraT5-tweet-small) |
26
+
27
+ # BibTex
28
+
29
+ If you use our models (Arat5-base, Arat5-msa-base, Arat5-tweet-base, Arat5-msa-small, or Arat5-tweet-small ) for your scientific publication, or if you find the resources in this repository useful, please cite our paper as follows (to be updated):
30
+ ```bibtex
31
+ @inproceedings{araT5-2021,
32
+ title = "{AraT5: Text-to-Text Transformers for Arabic Language Understanding and Generation",
33
+ author = "Nagoudi, El Moatez Billah and
34
+ Elmadany, AbdelRahim and
35
+ Abdul-Mageed, Muhammad",
36
+ booktitle = "https://arxiv.org/abs/2109.12068",
37
+ month = aug,
38
+ year = "2021"}
39
+ ```
40
+
41
+ ## Acknowledgments
42
+ We gratefully acknowledge support from the Natural Sciences and Engineering Research Council of Canada, the Social Sciences and Humanities Research Council of Canada, Canadian Foundation for Innovation, [ComputeCanada](www.computecanada.ca) and [UBC ARC-Sockeye](https://doi.org/10.14288/SOCKEYE). We also thank the [Google TensorFlow Research Cloud (TFRC)](https://www.tensorflow.org/tfrc) program for providing us with free TPU access.