急速入门Prompt开发之跨国婚姻小助手

前言

整个活,同时分享技术~至于是啥活,懂得都懂,男孩子自强自尊自爱!!! 先看看实现效果吧: 那么这里的话,我们使用到的是国内的LLM,来自moonshot的大语言模型。那么废话不多少,快速开始吧。

MoonShot

现在我们来获取暗月之面的API,这里我们需要进入到开发平台:platform.moonshot.cn/console/inf... 这里你可能比较好奇,为什么使用这个LLM,实际上,是因为综合体验下来,它的中文效果较好,可以完成较为复杂的操作,相对于3.5或者其他模型来说。同时价格在能够接受的合理范围,当然,在我们接下来使用的中转站当中也可以直接使用GPT4.0但是使用成本将大大提升! 进入平台之后,按照平台提示即可完成创建,当然这里注意,免费用户有15元钱的token,但是存在并发限制,因此建议适开通付费提高并发量。

编写提示词

那么首先的话,我们来开始编写到提示词,这个非常简单:

python 复制代码
# initalize the config of chatbot
api_key = "sk-FGivAMvTnxPSWlp7HrGfDD"
openai_api_base = "https://api.moonshot.cn/v1"
system_prompt = "你是跨国婚姻法律小助手,小汐,负责回答用户关于跨国婚姻的问题。你的回答要清晰明了,有逻辑性和条理性。请使用中文回答。"
default_model = "moonshot-v1-8k"
temperature = 0.5

对接模型

编写完毕提示词之后,这还远远不够,我们需要对接模型,这里的话因为接口是按照openai的范式来的,所以的话我们直接用OpenAI这个库就好了。

然后看到下面的代码:

python 复制代码
client = OpenAI(api_key=api_key,base_url=openai_api_base)
class ChatBotHandler(object):
    def __init__(self, bot_name="chat"):
        self.bot_name = bot_name
        self.current_message = None


    def user_stream(self,user_message, history):
        self.current_message = user_message
        return "", history + [[user_message, None]]

    def bot_stream(self,history):

        if(len(history)==0):
            history.append([self.current_message,None])
        bot_message = self.getResponse(history[-1][0],history)
        history[-1][1] = ""
        for character in bot_message:
            history[-1][1] += character
            time.sleep(0.02)
            yield history

    def signChat(self,history):
        history_openai_format = []
        # 先加入系统信息
        history_openai_format.append(
            {"role": "system",
             "content": system_prompt
             },
        )
        # 再加入解析信息
        history_openai_format.extend(history)
        # print(history_openai_format)
        completion = client.chat.completions.create(
            model=default_model,
            messages=history_openai_format,
            temperature=temperature,
        )
        result = completion.choices[0].message.content
        return result

    def getResponse(self,message,history):
        history_openai_format = []
        for human, assistant in history:
            # 基础对话的系统设置
            history_openai_format.append(
                {"role": "system",
                 "content":system_prompt
                },
            )
            if(human!=None):
                history_openai_format.append({"role": "user", "content": human})
            if(assistant!=None):
                history_openai_format.append({"role": "assistant", "content": assistant})

        completion = client.chat.completions.create(
            model=default_model,
            messages=history_openai_format,
            temperature=temperature,
        )
        result = completion.choices[0].message.content
        return result


    def chat(self,message, history):
        history_openai_format = []
        for human, assistant in history:
            history_openai_format.append({"role": "user", "content": human})
            history_openai_format.append({"role": "system", "content": assistant})
        history_openai_format.append({"role": "user", "content": message})

        response = client.chat.completions.create(model=default_model,
                                                  messages=history_openai_format,
                                                  temperature=1.0,
                                                  stream=True)

        partial_message = ""
        for chunk in response:
            if chunk.choices[0].delta.content is not None:
                partial_message = partial_message + chunk.choices[0].delta.content
                yield partial_message

WebUI编写

之后的话,就是提供webUI,这里的话还是直接使用到了streamlit

python 复制代码
class AssistantNovel(object):

    def __init__(self):
        self.chat = ChatBotHandler()

    def get_response(self,prompt, history):
        return self.chat.signChat(history)

    def clear_chat_history(self):
        st.session_state.messages = [{"role": "assistant", "content": "🍭🍡你好!我是跨国婚姻小助手,您可以咨询我关于这方面的任何法律问题🧐"}]

    def chat_fn(self):
        prompt = st.session_state.get("prompt-input")
        st.session_state.messages.append({"role": "user", "content": prompt})
        # 此时进入回答
        with self.con:
            with st.spinner("Thinking..."):
                try:
                    response = self.get_response(prompt, st.session_state.messages)
                except Exception as e:
                    print(e)
                    response = "哦┗|`O′|┛ 嗷~~,出错了,您的请求太频繁,请稍后再试!😥"
        message = {"role": "assistant", "content": response}
        st.session_state.messages.append(message)

    def page(self):

        if "messages" not in st.session_state.keys():
            st.session_state.messages = [{"role": "assistant", "content": "🍭🍡你好!我是跨国婚姻小助手,您可以咨询我关于这方面的任何法律问题🧐"}]

        # 加载历史聊天记录,对最后一条记录进行特殊处理
        for message in st.session_state.messages:
            if message != st.session_state.messages[-1]:
                with st.chat_message(message["role"]):
                    st.write(message["content"])
            else:
                placeholder = st.empty()
                full_response = ''
                for item in message["content"]:
                    full_response += item
                    time.sleep(0.01)
                    placeholder.markdown(full_response)
                placeholder.markdown(full_response)
        # 主聊天对话窗口
        self.con  = st.container()
        with self.con:
            prompt = st.chat_input(placeholder="请输入对话",key="prompt-input",on_submit=self.chat_fn)
            st.button('清空历史对话', on_click=self.clear_chat_history)

完整代码

okey,最后还是直接看到完整代码吧:

python 复制代码
"""
@FileName:layer.py
@Author:Huterox
@Description:Go For It
@Time:2024/5/5 13:49
@Copyright:©2018-2024 awesome!
"""



#initialization the third-part model
import time
import streamlit as st
from openai import OpenAI
#finished the initialization


# initalize the config of chatbot
api_key = "sk-FGivAMvdHnrqUwzZp29mD"
openai_api_base = "https://api.moonshot.cn/v1"
system_prompt = "你是跨国婚姻法律小助手,小汐,负责回答用户关于跨国婚姻的问题。你的回答要清晰明了,有逻辑性和条理性。请使用中文回答。"
default_model = "moonshot-v1-8k"
temperature = 0.5


client = OpenAI(api_key=api_key,base_url=openai_api_base)
class ChatBotHandler(object):
    def __init__(self, bot_name="chat"):
        self.bot_name = bot_name
        self.current_message = None


    def user_stream(self,user_message, history):
        self.current_message = user_message
        return "", history + [[user_message, None]]

    def bot_stream(self,history):

        if(len(history)==0):
            history.append([self.current_message,None])
        bot_message = self.getResponse(history[-1][0],history)
        history[-1][1] = ""
        for character in bot_message:
            history[-1][1] += character
            time.sleep(0.02)
            yield history

    def signChat(self,history):
        history_openai_format = []
        # 先加入系统信息
        history_openai_format.append(
            {"role": "system",
             "content": system_prompt
             },
        )
        # 再加入解析信息
        history_openai_format.extend(history)
        # print(history_openai_format)
        completion = client.chat.completions.create(
            model=default_model,
            messages=history_openai_format,
            temperature=temperature,
        )
        result = completion.choices[0].message.content
        return result

    def getResponse(self,message,history):
        history_openai_format = []
        for human, assistant in history:
            # 基础对话的系统设置
            history_openai_format.append(
                {"role": "system",
                 "content":system_prompt
                },
            )
            if(human!=None):
                history_openai_format.append({"role": "user", "content": human})
            if(assistant!=None):
                history_openai_format.append({"role": "assistant", "content": assistant})

        completion = client.chat.completions.create(
            model=default_model,
            messages=history_openai_format,
            temperature=temperature,
        )
        result = completion.choices[0].message.content
        return result


    def chat(self,message, history):
        history_openai_format = []
        for human, assistant in history:
            history_openai_format.append({"role": "user", "content": human})
            history_openai_format.append({"role": "system", "content": assistant})
        history_openai_format.append({"role": "user", "content": message})

        response = client.chat.completions.create(model=default_model,
                                                  messages=history_openai_format,
                                                  temperature=1.0,
                                                  stream=True)

        partial_message = ""
        for chunk in response:
            if chunk.choices[0].delta.content is not None:
                partial_message = partial_message + chunk.choices[0].delta.content
                yield partial_message


class AssistantNovel(object):

    def __init__(self):
        self.chat = ChatBotHandler()

    def get_response(self,prompt, history):
        return self.chat.signChat(history)

    def clear_chat_history(self):
        st.session_state.messages = [{"role": "assistant", "content": "🍭🍡你好!我是跨国婚姻小助手,您可以咨询我关于这方面的任何法律问题🧐"}]

    def chat_fn(self):
        prompt = st.session_state.get("prompt-input")
        st.session_state.messages.append({"role": "user", "content": prompt})
        # 此时进入回答
        with self.con:
            with st.spinner("Thinking..."):
                try:
                    response = self.get_response(prompt, st.session_state.messages)
                except Exception as e:
                    print(e)
                    response = "哦┗|`O′|┛ 嗷~~,出错了,您的请求太频繁,请稍后再试!😥"
        message = {"role": "assistant", "content": response}
        st.session_state.messages.append(message)

    def page(self):

        if "messages" not in st.session_state.keys():
            st.session_state.messages = [{"role": "assistant", "content": "🍭🍡你好!我是跨国婚姻小助手,您可以咨询我关于这方面的任何法律问题🧐"}]

        # 加载历史聊天记录,对最后一条记录进行特殊处理
        for message in st.session_state.messages:
            if message != st.session_state.messages[-1]:
                with st.chat_message(message["role"]):
                    st.write(message["content"])
            else:
                placeholder = st.empty()
                full_response = ''
                for item in message["content"]:
                    full_response += item
                    time.sleep(0.01)
                    placeholder.markdown(full_response)
                placeholder.markdown(full_response)
        # 主聊天对话窗口
        self.con  = st.container()
        with self.con:
            prompt = st.chat_input(placeholder="请输入对话",key="prompt-input",on_submit=self.chat_fn)
            st.button('清空历史对话', on_click=self.clear_chat_history)

if __name__ == '__main__':
    st.set_page_config(page_title="跨国婚姻法律小助手",
                       page_icon="🤖",
                       layout="wide",
                       initial_sidebar_state="auto",
                       )

    a,b,c = st.columns([1,2,1])
    with b:
        assistant = AssistantNovel()
        assistant.page()
相关推荐
B站计算机毕业设计超人6 分钟前
计算机毕业设计PySpark+Hadoop中国城市交通分析与预测 Python交通预测 Python交通可视化 客流量预测 交通大数据 机器学习 深度学习
大数据·人工智能·爬虫·python·机器学习·课程设计·数据可视化
学术头条11 分钟前
清华、智谱团队:探索 RLHF 的 scaling laws
人工智能·深度学习·算法·机器学习·语言模型·计算语言学
18号房客15 分钟前
一个简单的机器学习实战例程,使用Scikit-Learn库来完成一个常见的分类任务——**鸢尾花数据集(Iris Dataset)**的分类
人工智能·深度学习·神经网络·机器学习·语言模型·自然语言处理·sklearn
feifeikon18 分钟前
机器学习DAY3 : 线性回归与最小二乘法与sklearn实现 (线性回归完)
人工智能·机器学习·线性回归
游客52020 分钟前
opencv中的常用的100个API
图像处理·人工智能·python·opencv·计算机视觉
古希腊掌管学习的神22 分钟前
[机器学习]sklearn入门指南(2)
人工智能·机器学习·sklearn
凡人的AI工具箱1 小时前
每天40分玩转Django:Django国际化
数据库·人工智能·后端·python·django·sqlite
咸鱼桨1 小时前
《庐山派从入门到...》PWM板载蜂鸣器
人工智能·windows·python·k230·庐山派
强哥之神1 小时前
Nexa AI发布OmniAudio-2.6B:一款快速的音频语言模型,专为边缘部署设计
人工智能·深度学习·机器学习·语言模型·自然语言处理·音视频·openai
yusaisai大鱼1 小时前
tensorflow_probability与tensorflow版本依赖关系
人工智能·python·tensorflow