init commit

README.md

@@ -18,9 +18,9 @@ widget:
 
 ## 简介 Brief Introduction
 
-在Randeng-T5-Char-57M的基础上，收集了100个左右的中文数据集，进行Text2Text统一范式的有监督任务预训练。
+在Randeng-T5-Char-57M-Chinese的基础上，收集了100个左右的中文数据集，进行Text2Text统一范式的有监督任务预训练。
 
-On the basis of Randeng-T5-Char-57M, about 100 Chinese datasets were collected and pre-trained for the supervised task of Text2Text unified paradigm.
+On the basis of Randeng-T5-Char-57M-Chinese, about 100 Chinese datasets were collected and pre-trained for the supervised task of Text2Text unified paradigm.
 
 ## 模型分类 Model Taxonomy
 
@@ -33,9 +33,9 @@ On the basis of Randeng-T5-Char-57M, about 100 Chinese datasets were collected a
 
 参考论文：[Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](http://jmlr.org/papers/v21/20-074.html)
 
-基于[Randeng-T5-Char-57M](https://huggingface.co/IDEA-CCNL/Randeng-T5-Char-57M)，我们在收集的100+个中文领域的多任务数据集（从中采样了30w+个样本）上微调了它，得到了此多任务版本。这些多任务包括：情感分析，新闻分类，文本分类，意图识别，自然语言推理，多项选择，指代消解，抽取式阅读理解，实体识别，关键词抽取，生成式摘要。
+基于[Randeng-T5-Char-57M-Chinese](https://huggingface.co/IDEA-CCNL/Randeng-T5-Char-57M-Chinese)，我们在收集的100+个中文领域的多任务数据集（从中采样了30w+个样本）上微调了它，得到了此多任务版本。这些多任务包括：情感分析，新闻分类，文本分类，意图识别，自然语言推理，多项选择，指代消解，抽取式阅读理解，实体识别，关键词抽取，生成式摘要。
 
-Based on [Randeng-T5-Char-57M](https://huggingface.co/IDEA-CCNL/Randeng-T5-Char-57M), we fine-tuned it on a collection of 100+ multitasking datasets in Chinese domains (from which 30w+ samples were sampled) to obtain this multitasking version. These multitasks include: sentiment analysis, news classification, text classification, intention recognition, natural language inference, multiple choice, denotational disambiguation, extractive reading comprehension, entity recognition, keyword extraction, and generative summarization.
+Based on [Randeng-T5-Char-57M-Chinese](https://huggingface.co/IDEA-CCNL/Randeng-T5-Char-57M-Chinese), we fine-tuned it on a collection of 100+ multitasking datasets in Chinese domains (from which 30w+ samples were sampled) to obtain this multitasking version. These multitasks include: sentiment analysis, news classification, text classification, intention recognition, natural language inference, multiple choice, denotational disambiguation, extractive reading comprehension, entity recognition, keyword extraction, and generative summarization.
 
 
 ## 使用 Usage
@@ -54,7 +54,7 @@ model.resize_token_embeddings(len(tokenizer))
 model.eval()
 
 # tokenize
-text = "情感分析任务：【房间还是比较舒适的,酒店服务良好】这篇文章的情感态度是什么？正面/负面评"
+text = "情感分析任务：【房间还是比较舒适的,酒店服务良好】这篇文章的情感态度是什么？正面/负面"
 encode_dict = tokenizer(text, max_length=512, padding='max_length',truncation=True)
 
 inputs = {