Instructions to use DDDSSS/translation_en-zh with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use DDDSSS/translation_en-zh with Transformers:
# 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="DDDSSS/translation_en-zh")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("DDDSSS/translation_en-zh", dtype="auto") - Notebooks
- Google Colab
- Kaggle
微调脚本不完整
#2
by hachman - opened
load_dataset,evaluate.load,training_args这些找不到,以及数据集格式未知,是否可以提供完整版供参考?
详看参考的相关数据DDDSSS/en-zh-dataset
已查看,可以请教如何从0开始训练这类翻译模型吗?是否有参考教程?
不好意思,忘记回复你了,你可以看一下 huggingface 的翻译那一章节,主要参考他们的微调方式 trainer,在模型卡片介绍的时候已经说明如何微调了。不建议从0开始训练模型,因为很难训练出像他们这样的模型,需要考虑的是文本语料的数据量和质量。一般来说都是用各大机构训练好的模型,进行微调,微调数据最好不少于100W条数据。