# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("facebook/mbart-large-cc25")
model = AutoModelForSeq2SeqLM.from_pretrained("facebook/mbart-large-cc25", device_map="auto")Quick Links
mbart-large-cc25
Pretrained (not finetuned) multilingual mbart model. Original Languages
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
Original Code: https://github.com/pytorch/fairseq/tree/master/examples/mbart Docs: https://huggingface.co/transformers/master/model_doc/mbart.html Finetuning Code: examples/seq2seq/finetune.py (as of Aug 20, 2020)
Can also be finetuned for summarization.
- Downloads last month
- 21,585
# Use a pipeline as a high-level helper # Warning: Pipeline type "translation" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("translation", model="facebook/mbart-large-cc25")