wissamantoun's picture
Update README.md
c594996
---
language: ar
datasets:
- wikipedia
- Osian
- 1.5B-Arabic-Corpus
- oscar-arabic-unshuffled
- Assafir(private)
widget:
- text: " عاصمة لبنان هي [MASK] ."
---
# AraBERT v1 & v2 : Pre-training BERT for Arabic Language Understanding
<img src="https://raw.githubusercontent.com/aub-mind/arabert/master/arabert_logo.png" width="100" align="left"/>
**AraBERT** is an Arabic pretrained language model based on [Google's BERT architechture](https://github.com/google-research/bert). AraBERT uses the same BERT-Base config. More details are available in the [AraBERT Paper](https://arxiv.org/abs/2003.00104) and in the [AraBERT Meetup](https://github.com/WissamAntoun/pydata_khobar_meetup)
There are two versions of the model, AraBERTv0.1 and AraBERTv1, with the difference being that AraBERTv1 uses pre-segmented text where prefixes and suffixes were split using the [Farasa Segmenter](http://alt.qcri.org/farasa/segmenter.html).
We evaluate AraBERT models on different downstream tasks and compare them to [mBERT]((https://github.com/google-research/bert/blob/master/multilingual.md)), and other state of the art models (*To the extent of our knowledge*). The Tasks were Sentiment Analysis on 6 different datasets ([HARD](https://github.com/elnagara/HARD-Arabic-Dataset), [ASTD-Balanced](https://www.aclweb.org/anthology/D15-1299), [ArsenTD-Lev](https://staff.aub.edu.lb/~we07/Publications/ArSentD-LEV_Sentiment_Corpus.pdf), [LABR](https://github.com/mohamedadaly/LABR)), Named Entity Recognition with the [ANERcorp](http://curtis.ml.cmu.edu/w/courses/index.php/ANERcorp), and Arabic Question Answering on [Arabic-SQuAD and ARCD](https://github.com/husseinmozannar/SOQAL)
# AraBERTv2
## What's New!
AraBERT now comes in 4 new variants to replace the old v1 versions:
More Detail in the AraBERT folder and in the [README](https://github.com/aub-mind/arabert/blob/master/AraBERT/README.md) and in the [AraBERT Paper](https://arxiv.org/abs/2003.00104v2)
Model | HuggingFace Model Name | Size (MB/Params)| Pre-Segmentation | DataSet (Sentences/Size/nWords) |
---|:---:|:---:|:---:|:---:
AraBERTv0.2-base | [bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) | 543MB / 136M | No | 200M / 77GB / 8.6B |
AraBERTv0.2-large| [bert-large-arabertv02](https://huggingface.co/aubmindlab/bert-large-arabertv02) | 1.38G 371M | No | 200M / 77GB / 8.6B |
AraBERTv2-base| [bert-base-arabertv2](https://huggingface.co/aubmindlab/bert-base-arabertv2) | 543MB 136M | Yes | 200M / 77GB / 8.6B |
AraBERTv2-large| [bert-large-arabertv2](https://huggingface.co/aubmindlab/bert-large-arabertv2) | 1.38G 371M | Yes | 200M / 77GB / 8.6B |
AraBERTv0.2-Twitter-base| [bert-base-arabertv02-twitter](https://huggingface.co/aubmindlab/bert-base-arabertv02-twitter) | 543MB / 136M | No | Same as v02 + 60M Multi-Dialect Tweets|
AraBERTv0.2-Twitter-large| [bert-large-arabertv02-twitter](https://huggingface.co/aubmindlab/bert-large-arabertv02-twitter) | 1.38G / 371M | No | Same as v02 + 60M Multi-Dialect Tweets|
AraBERTv0.1-base| [bert-base-arabertv01](https://huggingface.co/aubmindlab/bert-base-arabertv01) | 543MB 136M | No | 77M / 23GB / 2.7B |
AraBERTv1-base| [bert-base-arabert](https://huggingface.co/aubmindlab/bert-base-arabert) | 543MB 136M | Yes | 77M / 23GB / 2.7B |
All models are available in the `HuggingFace` model page under the [aubmindlab](https://huggingface.co/aubmindlab/) name. Checkpoints are available in PyTorch, TF2 and TF1 formats.
## Better Pre-Processing and New Vocab
We identified an issue with AraBERTv1's wordpiece vocabulary. The issue came from punctuations and numbers that were still attached to words when learned the wordpiece vocab. We now insert a space between numbers and characters and around punctuation characters.
The new vocabulary was learned using the `BertWordpieceTokenizer` from the `tokenizers` library, and should now support the Fast tokenizer implementation from the `transformers` library.
**P.S.**: All the old BERT codes should work with the new BERT, just change the model name and check the new preprocessing function
**Please read the section on how to use the [preprocessing function](#Preprocessing)**
## Bigger Dataset and More Compute
We used ~3.5 times more data, and trained for longer.
For Dataset Sources see the [Dataset Section](#Dataset)
Model | Hardware | num of examples with seq len (128 / 512) |128 (Batch Size/ Num of Steps) | 512 (Batch Size/ Num of Steps) | Total Steps | Total Time (in Days) |
---|:---:|:---:|:---:|:---:|:---:|:---:
AraBERTv0.2-base | TPUv3-8 | 420M / 207M | 2560 / 1M | 384/ 2M | 3M | -
AraBERTv0.2-large | TPUv3-128 | 420M / 207M | 13440 / 250K | 2056 / 300K | 550K | 7
AraBERTv2-base | TPUv3-8 | 420M / 207M | 2560 / 1M | 384/ 2M | 3M | -
AraBERTv2-large | TPUv3-128 | 520M / 245M | 13440 / 250K | 2056 / 300K | 550K | 7
AraBERT-base (v1/v0.1) | TPUv2-8 | - |512 / 900K | 128 / 300K| 1.2M | 4
# Dataset
The pretraining data used for the new AraBERT model is also used for Arabic **GPT2 and ELECTRA**.
The dataset consists of 77GB or 200,095,961 lines or 8,655,948,860 words or 82,232,988,358 chars (before applying Farasa Segmentation)
For the new dataset we added the unshuffled OSCAR corpus, after we thoroughly filter it, to the previous dataset used in AraBERTv1 but with out the websites that we previously crawled:
- OSCAR unshuffled and filtered.
- [Arabic Wikipedia dump](https://archive.org/details/arwiki-20190201) from 2020/09/01
- [The 1.5B words Arabic Corpus](https://www.semanticscholar.org/paper/1.5-billion-words-Arabic-Corpus-El-Khair/f3eeef4afb81223df96575adadf808fe7fe440b4)
- [The OSIAN Corpus](https://www.aclweb.org/anthology/W19-4619)
- Assafir news articles. Huge thank you for Assafir for providing us the data
# Preprocessing
It is recommended to apply our preprocessing function before training/testing on any dataset.
**Install the arabert python package to segment text for AraBERT v1 & v2 or to clean your data `pip install arabert`**
```python
from arabert.preprocess import ArabertPreprocessor
model_name="aubmindlab/bert-large-arabertv02"
arabert_prep = ArabertPreprocessor(model_name=model_name)
text = "ولن نبالغ إذا قلنا: إن هاتف أو كمبيوتر المكتب في زمننا هذا ضروري"
arabert_prep.preprocess(text)
>>> output: ولن نبالغ إذا قلنا : إن هاتف أو كمبيوتر المكتب في زمننا هذا ضروري
```
# TensorFlow 1.x models
The TF1.x model are available in the HuggingFace models repo.
You can download them as follows:
- via git-lfs: clone all the models in a repo
```bash
curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | sudo bash
sudo apt-get install git-lfs
git lfs install
git clone https://huggingface.co/aubmindlab/MODEL_NAME
tar -C ./MODEL_NAME -zxvf /content/MODEL_NAME/tf1_model.tar.gz
```
where `MODEL_NAME` is any model under the `aubmindlab` name
- via `wget`:
- Go to the tf1_model.tar.gz file on huggingface.co/models/aubmindlab/MODEL_NAME.
- copy the `oid sha256`
- then run `wget https://cdn-lfs.huggingface.co/aubmindlab/aragpt2-base/INSERT_THE_SHA_HERE` (ex: for `aragpt2-base`: `wget https://cdn-lfs.huggingface.co/aubmindlab/aragpt2-base/3766fc03d7c2593ff2fb991d275e96b81b0ecb2098b71ff315611d052ce65248`)
# If you used this model please cite us as :
Google Scholar has our Bibtex wrong (missing name), use this instead
```
@inproceedings{antoun2020arabert,
title={AraBERT: Transformer-based Model for Arabic Language Understanding},
author={Antoun, Wissam and Baly, Fady and Hajj, Hazem},
booktitle={LREC 2020 Workshop Language Resources and Evaluation Conference 11--16 May 2020},
pages={9}
}
```
# Acknowledgments
Thanks to TensorFlow Research Cloud (TFRC) for the free access to Cloud TPUs, couldn't have done it without this program, and to the [AUB MIND Lab](https://sites.aub.edu.lb/mindlab/) Members for the continuous support. Also thanks to [Yakshof](https://www.yakshof.com/#/) and Assafir for data and storage access. Another thanks for Habib Rahal (https://www.behance.net/rahalhabib), for putting a face to AraBERT.
# Contacts
**Wissam Antoun**: [Linkedin](https://www.linkedin.com/in/wissam-antoun-622142b4/) | [Twitter](https://twitter.com/wissam_antoun) | [Github](https://github.com/WissamAntoun) | <wfa07@mail.aub.edu> | <wissam.antoun@gmail.com>
**Fady Baly**: [Linkedin](https://www.linkedin.com/in/fadybaly/) | [Twitter](https://twitter.com/fadybaly) | [Github](https://github.com/fadybaly) | <fgb06@mail.aub.edu> | <baly.fady@gmail.com>