Transformers
PyTorch
Safetensors
Chinese
t5
text2text-generation
Text2Text Generation
T5
chinese
text-generation-inference
Instructions to use IDEA-CCNL/Randeng-T5-Char-700M-MultiTask-Chinese with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use IDEA-CCNL/Randeng-T5-Char-700M-MultiTask-Chinese with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("IDEA-CCNL/Randeng-T5-Char-700M-MultiTask-Chinese") model = AutoModelForSeq2SeqLM.from_pretrained("IDEA-CCNL/Randeng-T5-Char-700M-MultiTask-Chinese") - Notebooks
- Google Colab
- Kaggle
这个char的效果比另一个不带char的差好多呀
#1
by awdrgyjilplij - opened
我主要试了阅读理解任务,这个大概f1有50,另一个f1能有65。
分词方式能带来这么大的差别吗
我主要试了阅读理解任务,这个大概f1有50,另一个f1能有65。
分词方式能带来这么大的差别吗
我们尝试了一些任务,一般来说生成类任务sentence-piece会比char-level的表现好不少