Arabertmo Base V6

🚀 Arabic BERT Model

AraBERTMo is an Arabic pre - trained language model based on Google's BERT architecture, offering various variants and achieving good results in fill - mask tasks.

🚀 Quick Start

You can use this model by installing torch or tensorflow and Huggingface library transformers. And you can use it directly by initializing it like this:

from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("Ebtihal/AraBertMo_base_V6")
model = AutoModelForMaskedLM.from_pretrained("Ebtihal/AraBertMo_base_V6")

✨ Features

• Multiple Variants: AraBERTMo_base now comes in 10 new variants.
• Available on HuggingFace: All models are available on the HuggingFace model page under the Ebtihal name.
• PyTorch Format: Checkpoints are available in PyTorch formats.

📦 Installation

The installation mainly involves installing torch or tensorflow and the Huggingface library transformers.

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("Ebtihal/AraBertMo_base_V6")
model = AutoModelForMaskedLM.from_pretrained("Ebtihal/AraBertMo_base_V6")

📚 Documentation

Pretraining Corpus

AraBertMo_base_V6 model was pre - trained on ~3 million words:

• OSCAR - Arabic version "unshuffled_deduplicated_ar".

Training results

This model achieves the following results:

Model Background

This model was built for master's degree research in an organization:

• University of kufa.
• Faculty of Computer Science and Mathematics.
• Department of Computer Science

Additional Information

• Language: Arabic
• Tags: Fill - Mask
• Datasets: OSCAR
• Widget Examples:
  • " السلام عليكم ورحمة[MASK] وبركاتة"
  • " اهلا وسهلا بكم في [MASK] من سيربح المليون"
  • " مرحبا بك عزيزي الزائر [MASK] موقعنا "