Mbart Large Cc25

🚀 mbart-large-cc25

A pretrained (not finetuned) multilingual mbart model.

🚀 Quick Start

This is a pretrained multilingual mbart model that has not been finetuned. It supports multiple languages including English, Arabic, Czech, German, Estonian, Finnish, French, Gujarati, Hindi, Italian, Japanese, Kazakh, Korean, Lithuanian, Latvian, Myanmar, Nepali, Dutch, Romanian, Russian, Sinhala, Turkish, Vietnamese, Chinese, and multilingual settings.

✨ Features

• Multilingual Support: It can handle a wide range of languages, as specified in the tags.
• Finetuning Capability: Can be finetuned for tasks such as summarization.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

The original languages supported by the model are defined as follows:

export langs=ar_AR,cs_CZ,de_DE,en_XX,es_XX,et_EE,fi_FI,fr_XX,gu_IN,hi_IN,it_IT,ja_XX,kk_KZ,ko_KR,lt_LT,lv_LV,my_MM,ne_NP,nl_XX,ro_RO,ru_RU,si_LK,tr_TR,vi_VN,zh_CN

📚 Documentation

• Original Code: You can find the original code at https://github.com/pytorch/fairseq/tree/master/examples/mbart.
• Model Documentation: Refer to the official documentation at https://huggingface.co/transformers/master/model_doc/mbart.html.
• Finetuning Code: As of Aug 20, 2020, the finetuning code can be found at examples/seq2seq/finetune.py.

This model can also be finetuned for summarization tasks.