monsoon-nlp/sanaa

This is a small GPT-2 model retrained on Arabic Wikipedia circa September 2020 . There is NO content filtering in the current version; do not use for public-facing text generation . I finetuned this model on different Arabic dialects to generate a new model (monsoon-nlp/sanaa-dialect on HuggingFace) with some additional .control tokens . The model was very small in my project and would be improved by adding many X more training data . [Sanaa]Arabic GPT2 demo demonstrates how to train a GPT model using TensorFlow and generate text in SimpleTransformers. The model is based on the first 600,,