FEAT: first commit

README.md

@@ -10,7 +10,7 @@ widget:
 
 
 ---
-# Randeng-BART-759M-BertTokenizer model (Chinese)，one model of [Fengshenbang-LM](https://github.com/IDEA-CCNL/Fengshenbang-LM)
+# Randeng-BART-759M-Chinese-BertTokenizer model (Chinese)，one model of [Fengshenbang-LM](https://github.com/IDEA-CCNL/Fengshenbang-LM)
 
 The 759M million parameter Randeng-BART large model, using 180G Chinese data, 8 A100(40G) training for 7 days，which is a Encoder-Only transformer structure.
 
@@ -18,7 +18,7 @@ We use bert vocab as our tokenizer.
 
 ## Task Description
 
-Randeng-BART-759M-BertTokenizer is pre-trained by Text-Infilling task from BART [paper](https://readpaper.com/pdf-annotate/note?noteId=675945911766249472&pdfId=550970997159968917)
+Randeng-BART-759M-Chinese-BertTokenizer is pre-trained by Text-Infilling task from BART [paper](https://readpaper.com/pdf-annotate/note?noteId=675945911766249472&pdfId=550970997159968917)
 
 You can find our pretrain's code in [Fengshengbang-LM](https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/pretrain_randeng_bart)
 
@@ -28,8 +28,8 @@ You can find our pretrain's code in [Fengshengbang-LM](https://github.com/IDEA-C
 from transformers import BartForConditionalGeneration, AutoTokenizer, Text2TextGenerationPipeline
 import torch
 
-tokenizer=AutoTokenizer.from_pretrained('IDEA-CCNL/Randeng-BART-759M-BertTokenizer', use_fast=false)
-model=BartForConditionalGeneration.from_pretrained('IDEA-CCNL/Randeng-BART-759M-BertTokenizer')
+tokenizer=AutoTokenizer.from_pretrained('IDEA-CCNL/Randeng-BART-759M-Chinese-BertTokenizer', use_fast=false)
+model=BartForConditionalGeneration.from_pretrained('IDEA-CCNL/Randeng-BART-759M-Chinese-BertTokenizer')
 text = '桂林是著名的[MASK]，它有很多[MASK]。'
 text2text_generator = Text2TextGenerationPipeline(model, tokenizer)
 print(text2text_generator(text, max_length=50, do_sample=False))