Gpt2 Medium Italian Embeddings

🚀 GPT-2 recycled for Italian (medium, adapted lexical embeddings)

This model is based on the OpenAI GPT-2 medium model, with retrained lexical layers for an Italian vocabulary, enabling text generation in Italian.

🚀 Quick Start

This model allows you to generate Italian text. You can use it through the following code examples.

💻 Usage Examples

Basic Usage

from transformers import pipeline

pipe = pipeline("text-generation", model="GroNLP/gpt2-medium-italian-embeddings")

Advanced Usage

from transformers import AutoTokenizer, AutoModel, TFAutoModel

tokenizer = AutoTokenizer.from_pretrained("GroNLP/gpt2-medium-italian-embeddings")
model = AutoModel.from_pretrained("GroNLP/gpt2-medium-italian-embeddings")  # PyTorch
model = TFAutoModel.from_pretrained("GroNLP/gpt2-medium-italian-embeddings")  # Tensorflow

📚 Documentation

Model description

This model is based on the medium OpenAI GPT-2 (gpt2-medium) model. The Transformer layer weights in this model are identical to the original English model, but the lexical layer has been retrained for an Italian vocabulary. For details, check out our paper on arXiv and the code on Github.

Related models

Dutch

• gpt2-small-dutch-embeddings: Small model size with only retrained lexical embeddings.
• gpt2-small-dutch: Small model size with retrained lexical embeddings and additional fine - tuning of the full model. (Recommended)
• gpt2-medium-dutch-embeddings: Medium model size with only retrained lexical embeddings.

Italian

• gpt2-small-italian-embeddings: Small model size with only retrained lexical embeddings.
• gpt2-small-italian: Small model size with retrained lexical embeddings and additional fine - tuning of the full model. (Recommended)
• gpt2-medium-italian-embeddings: Medium model size with only retrained lexical embeddings.

📄 License

BibTeX entry

@misc{devries2020good,
      title={As good as new. How to successfully recycle English GPT-2 to make models for other languages}, 
      author={Wietse de Vries and Malvina Nissim},
      year={2020},
      eprint={2012.05628},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}