Bert2bert L 24 Wmt En De

🚀 bert2bert_L-24_wmt_en_de EncoderDecoder model

This is an encoder-decoder model designed for English to German translation, initialized on bert-large checkpoints and fine-tuned on the WMT dataset.

The model was introduced in this paper by Sascha Rothe, Shashi Narayan, Aliaksei Severyn and first released in this repository. It is initialized on the bert-large checkpoints for both the encoder and decoder and fine-tuned on English to German translation using the WMT dataset.

Disclaimer: The model card has been written by the Hugging Face team.

🚀 Quick Start

✨ Features

• An encoder-decoder model for English to German translation.
• Initialized on bert-large checkpoints and fine-tuned on the WMT dataset.

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("google/bert2bert_L-24_wmt_en_de", pad_token="<pad>", eos_token="</s>", bos_token="<s>")
model = AutoModelForSeq2SeqLM.from_pretrained("google/bert2bert_L-24_wmt_en_de")

sentence = "Would you like to grab a coffee with me this week?"

input_ids = tokenizer(sentence, return_tensors="pt", add_special_tokens=False).input_ids
output_ids = model.generate(input_ids)[0]
print(tokenizer.decode(output_ids, skip_special_tokens=True))
# should output
# Möchten Sie diese Woche einen Kaffee mit mir schnappen?

📄 License

This model is licensed under the Apache-2.0 license.