app.py

import os
# install torch and tf
os.system('pip install transformers SentencePiece')
os.system('pip install torch')

from transformers import T5Tokenizer, T5ForConditionalGeneration, AutoTokenizer
import torch
import gradio as gr

# 下载模型
tokenizer = T5Tokenizer.from_pretrained("ClueAI/ChatYuan-large-v1")
model = T5ForConditionalGeneration.from_pretrained("ClueAI/ChatYuan-large-v1")
# 修改colab笔记本设置为gpu，推理更快
device = torch.device('cpu')
model.to(device)
print('Model Load done!')

def preprocess(text):
    text = text.replace("\n", "\\n").replace("\t", "\\t")
    return text

def postprocess(text):
    return text.replace("\\n", "\n").replace("\\t", "\t")

def answer(text, sample=True, top_p=0.9, temperature=0.7):
    '''sample：是否抽样。生成任务，可以设置为True;
    top_p：0-1之间，生成的内容越多样
    max_new_tokens=512 lost...'''
    text = preprocess(text)
    print('用户: '+text)
    encoding = tokenizer(text=[text], truncation=True, padding=True, max_length=768, return_tensors="pt").to(device) 
    if not sample:
        out = model.generate(**encoding, return_dict_in_generate=True, output_scores=False, max_new_tokens=512, num_beams=1, length_penalty=0.6)
    else:
        out = model.generate(**encoding, return_dict_in_generate=True, output_scores=False, max_new_tokens=512, do_sample=True, top_p=top_p, temperature=temperature, no_repeat_ngram_size=3)
    out_text = tokenizer.batch_decode(out["sequences"], skip_special_tokens=True)
    print('小元: '+postprocess(out_text[0]))
    return postprocess(out_text[0])

def command_result(text):
	output = answer(text)
	return output

iface = gr.Interface(fn=command_result, inputs="text", outputs="text", title="中文聊天机器人Demo")
iface.launch()