吴恩达 GPT Prompting 课程
Prompting Guidelines 目录
- Guidelines for Prompting
- Prompting Principles
- Principle 1: Write clear and specific instructions
- 1.1: Use delimiters to clearly indicate distinct parts of the input
- 1.2: Ask for a structured output
- 1.3: Ask the model to check whether conditions are satisfied
- 1.4: "Few-shot" prompting
- Principle 2: Give the model time to “think”
- 2.1.1: Specify the steps required to complete a task
- 2.1.2 Ask for output in a specified format
- 2.2: Instruct the model to work out its own solution before rushing to a conclusion
- Model Limitations: Hallucinations
Guidelines for Prompting
In this lesson, you’ll practice two prompting principles and their related tactics in order to write effective prompts for large language models.
Prompting Principles
- Principle 1: Write clear and specific instructions
- Use delimiters to clearly indicate distinct parts of the input
- Ask for a structured output
- Ask the model to check whether conditions are satisfied
- “Few-shot” prompting
- Principle 2: Give the model time to “think”
- Specify the steps required to complete a task
- Ask for output in a specified format
- Instruct the model to work out its own solution before rushing to a conclusion
Principle 1: Write clear and specific instructions
1.1: Use delimiters to clearly indicate distinct parts of the input
- Delimiters can be anything like:
```, """, < >, <tag> </tag>, :
text = f"""
You should express what you want a model to do by \
providing instructions that are as clear and \
specific as you can possibly make them. \
This will guide the model towards the desired output, \
and reduce the chances of receiving irrelevant \
or incorrect responses. Don't confuse writing a \
clear prompt with writing a short prompt. \
In many cases, longer prompts provide more clarity \
and context for the model, which can lead to \
more detailed and relevant outputs.
"""
prompt = f"""
Summarize the text delimited by triple backticks \
into a single sentence.
```{text}```
"""
response = get_completion(prompt)
print(response)
GPT 返回结果:
To guide a model towards the desired output and reduce irrelevant or incorrect responses, it is important to provide clear and specific instructions, which can be achieved through longer prompts that offer more clarity and context.
1.2: Ask for a structured output
- JSON, HTML
prompt = f"""
Generate a list of three made-up book titles along \
with their authors and genres.
Provide them in JSON format with the following keys:
book_id, title, author, genre.
"""
response = get_completion(prompt)
print(response)
GPT 返回结果:
{"books": [{"book_id": 1,"title": "The Enigma of Elysium","author": "Evelyn Sinclair","genre": "Mystery"},{"book_id": 2,"title": "Whispers in the Wind","author": "Nathaniel Blackwood","genre": "Fantasy"},{"book_id": 3,"title": "Echoes of the Past","author": "Amelia Hart","genre": "Romance"}]
}
1.3: Ask the model to check whether conditions are satisfied
text_1 = f"""
Making a cup of tea is easy! First, you need to get some \
water boiling. While that's happening, \
grab a cup and put a tea bag in it. Once the water is \
hot enough, just pour it over the tea bag. \
Let it sit for a bit so the tea can steep. After a \
few minutes, take out the tea bag. If you \
like, you can add some sugar or milk to taste. \
And that's it! You've got yourself a delicious \
cup of tea to enjoy.
"""
prompt = f"""
You will be provided with text delimited by triple quotes.
If it contains a sequence of instructions, \
re-write those instructions in the following format:Step 1 - ...
Step 2 - …
…
Step N - …If the text does not contain a sequence of instructions, \
then simply write \"No steps provided.\"\"\"\"{text_1}\"\"\"
"""
response = get_completion(prompt)
print("Completion for Text 1:")
print(response)
GPT 返回结果:
Completion for Text 1:
Step 1 - Get some water boiling.
Step 2 - Grab a cup and put a tea bag in it.
Step 3 - Once the water is hot enough, pour it over the tea bag.
Step 4 - Let it sit for a bit so the tea can steep.
Step 5 - After a few minutes, take out the tea bag.
Step 6 - If you like, add some sugar or milk to taste.
Step 7 - Enjoy your delicious cup of tea.
text_2 = f"""
The sun is shining brightly today, and the birds are \
singing. It's a beautiful day to go for a \
walk in the park. The flowers are blooming, and the \
trees are swaying gently in the breeze. People \
are out and about, enjoying the lovely weather. \
Some are having picnics, while others are playing \
games or simply relaxing on the grass. It's a \
perfect day to spend time outdoors and appreciate the \
beauty of nature.
"""
prompt = f"""
You will be provided with text delimited by triple quotes.
If it contains a sequence of instructions, \
re-write those instructions in the following format:Step 1 - ...
Step 2 - …
…
Step N - …If the text does not contain a sequence of instructions, \
then simply write \"No steps provided.\"\"\"\"{text_2}\"\"\"
"""
response = get_completion(prompt)
print("Completion for Text 2:")
print(response)
GPT 返回结果:
Completion for Text 2:
No steps provided.
1.4: “Few-shot” prompting
prompt = f"""
Your task is to answer in a consistent style.<child>: Teach me about patience.<grandparent>: The river that carves the deepest \
valley flows from a modest spring; the \
grandest symphony originates from a single note; \
the most intricate tapestry begins with a solitary thread.<child>: Teach me about resilience.
"""
response = get_completion(prompt)
print(response)
GPT 返回结果:
: Resilience is like a mighty oak tree that withstands the strongest storms, bending but never breaking. It is the unwavering determination to rise again after every fall, and the ability to find strength in the face of adversity. Just as a diamond is formed under immense pressure, resilience is forged through challenges and hardships, making us stronger and more resilient in the process.
Principle 2: Give the model time to “think”
2.1.1: Specify the steps required to complete a task
text = f"""
In a charming village, siblings Jack and Jill set out on \
a quest to fetch water from a hilltop \
well. As they climbed, singing joyfully, misfortune \
struck—Jack tripped on a stone and tumbled \
down the hill, with Jill following suit. \
Though slightly battered, the pair returned home to \
comforting embraces. Despite the mishap, \
their adventurous spirits remained undimmed, and they \
continued exploring with delight.
"""prompt_1 = f"""
Perform the following actions:
1 - Summarize the following text delimited by triple \
backticks with 1 sentence.
2 - Translate the summary into French.
3 - List each name in the French summary.
4 - Output a json object that contains the following \
keys: french_summary, num_names.Separate your answers with line breaks.Text:
```{text}```
"""
response = get_completion(prompt_1)
print("Completion for prompt 1:")
print(response)
GPT 返回结果:
Completion for prompt 1:
1 - Jack and Jill, siblings, go on a quest to fetch water from a hilltop well, but encounter misfortune when Jack trips on a stone and tumbles down the hill, with Jill following suit, yet they return home and remain undeterred in their adventurous spirits.2 - Jack et Jill, frère et sœur, partent en quête d'eau d'un puits au sommet d'une colline, mais rencontrent un malheur lorsque Jack trébuche sur une pierre et dévale la colline, suivi de Jill, pourtant ils rentrent chez eux et restent déterminés dans leur esprit d'aventure.3 - Jack, Jill4 - {"french_summary": "Jack et Jill, frère et sœur, partent en quête d'eau d'un puits au sommet d'une colline, mais rencontrent un malheur lorsque Jack trébuche sur une pierre et dévale la colline, suivi de Jill, pourtant ils rentrent chez eux et restent déterminés dans leur esprit d'aventure.","num_names": 2
}
2.1.2 Ask for output in a specified format
prompt_2 = f"""
Your task is to perform the following actions:
1 - Summarize the following text delimited by <> with 1 sentence.
2 - Translate the summary into French.
3 - List each name in the French summary.
4 - Output a json object that contains the following keys: french_summary, num_names.Use the following format:
Text: <text to summarize>
Summary: <summary>
Translation: <summary translation>
Names: <list of names in Italian summary>
Output JSON: <json with summary and num_names>Text: <{text}>
"""
response = get_completion(prompt_2)
print("\nCompletion for prompt 2:")
print(response)
GPT 返回结果:
Completion for prompt 2:
Summary: Jack and Jill, siblings, go on a quest to fetch water from a hilltop well but encounter misfortune along the way.
Translation: Jack et Jill, frère et sœur, partent en quête d'eau d'un puits au sommet d'une colline mais rencontrent des malheurs en chemin.
Names: Jack, Jill
Output JSON: {"french_summary": "Jack et Jill, frère et sœur, partent en quête d'eau d'un puits au sommet d'une colline mais rencontrent des malheurs en chemin.", "num_names": 2}
2.2: Instruct the model to work out its own solution before rushing to a conclusion
prompt = f"""
Determine if the student's solution is correct or not.Question:
I'm building a solar power installation and I need \help working out the financials.
- Land costs $100 / square foot
- I can buy solar panels for $250 / square foot
- I negotiated a contract for maintenance that will cost \
me a flat $100k per year, and an additional $10 / square \
foot
What is the total cost for the first year of operations
as a function of the number of square feet.Student's Solution:
Let x be the size of the installation in square feet.
Costs:
1. Land cost: 100x
2. Solar panel cost: 250x
3. Maintenance cost: 100,000 + 100x
Total cost: 100x + 250x + 100,000 + 100x = 450x + 100,000
"""
response = get_completion(prompt)
print(response)
GPT 返回结果:
The student's solution is correct. They correctly identified the costs for land, solar panels, and maintenance, and calculated the total cost as a function of the number of square feet.
Note that the student’s solution is actually not correct.
We can fix this by instructing the model to work out its own solution first (or not).
prompt = f"""
Your task is to determine if the student's solution \
is correct or not.
To solve the problem do the following:
- First, work out your own solution to the problem.
- Then compare your solution to the student's solution \
and evaluate if the student's solution is correct or not.
Don't decide if the student's solution is correct until
you have done the problem yourself.Use the following format:
Question:question hereStudent's solution:
student's solution hereActual solution:steps to work out the solution and your solution hereIs the student's solution the same as actual solution \
just calculated:yes or noStudent grade:correct or incorrectQuestion:I'm building a solar power installation and I need help \
working out the financials.
- Land costs $100 / square foot
- I can buy solar panels for $250 / square foot
- I negotiated a contract for maintenance that will cost \
me a flat $100k per year, and an additional $10 / square \
foot
What is the total cost for the first year of operations \
as a function of the number of square feet.Let x be the size of the installation in square feet.
Costs:
1. Land cost: 100x
2. Solar panel cost: 250x
3. Maintenance cost: 100,000 + 100x
Total cost: 100x + 250x + 100,000 + 100x = 450x + 100,000Actual solution:response = get_completion(prompt)
print(response)
GPT 返回结果:
To calculate the total cost for the first year of operations, we need to add up the costs of land, solar panels, and maintenance.Let x be the size of the installation in square feet.Costs:
1. Land cost: $100 * x
2. Solar panel cost: $250 * x
3. Maintenance cost: $100,000 + $10 * xTotal cost: $100 * x + $250 * x + $100,000 + $10 * x = $360 * x + $100,000Is the student's solution the same as the actual solution just calculated:
NoStudent grade:
Incorrect
Model Limitations: Hallucinations
- Boie is a real company, the product name is not real.
prompt = f"""
Tell me about AeroGlide UltraSlim Smart Toothbrush by Boie
"""
response = get_completion(prompt)
print(response)
GPT 返回结果:
The AeroGlide UltraSlim Smart Toothbrush by Boie is a technologically advanced toothbrush designed to provide a superior brushing experience. Boie is a company known for its innovative oral care products, and the AeroGlide UltraSlim Smart Toothbrush is no exception.One of the standout features of this toothbrush is its ultra-slim design. The brush head is only 2mm thick, making it much thinner than traditional toothbrushes. This slim profile allows for better access to hard-to-reach areas of the mouth, ensuring a thorough and effective clean.The AeroGlide UltraSlim Smart Toothbrush also incorporates smart technology. It connects to a mobile app via Bluetooth, allowing users to track their brushing habits and receive personalized recommendations for improving their oral hygiene routine. The app provides real-time feedback on brushing technique, ensuring that users are brushing for the recommended two minutes and covering all areas of their mouth.The toothbrush itself is made from a durable and hygienic material called thermoplastic elastomer. This material is non-porous, meaning it doesn't harbor bacteria or mold, making it more hygienic than traditional toothbrush bristles. The bristles are also ultra-soft, providing a gentle yet effective clean without causing any damage to the gums or enamel.In addition to its technological features, the AeroGlide UltraSlim Smart Toothbrush is also eco-friendly. It is designed to last for up to six months, reducing the need for frequent replacements. The brush head is also replaceable, further reducing waste.Overall, the AeroGlide UltraSlim Smart Toothbrush by Boie offers a combination of advanced technology, slim design, and eco-friendly features. It aims to provide users with a superior brushing experience, helping them maintain optimal oral health.
相关文章:
吴恩达 GPT Prompting 课程
Prompting Guidelines 目录 Guidelines for PromptingPrompting Principles Principle 1: Write clear and specific instructions1.1: Use delimiters to clearly indicate distinct parts of the input1.2: Ask for a structured output1.3: Ask the model to check whether …...
gpt3.5写MATLAB代码剪辑视频,使之保留画面ROI区域
% 输入和输出文件名 inputVideoFile input_video.mp4; outputVideoFile output_video.mp4;% 创建 VideoReader 和 VideoWriter 对象 videoReader VideoReader(inputVideoFile); outputVideo VideoWriter(outputVideoFile, MPEG-4); outputVideo.FrameRate videoReader.Fra…...
设计模式二十一:状态模式(State Pattern)
一个对象的内部状态发生变化时,允许对象改变其行为。这种模式使得一个对象看起来好像在运行时改变了它的类,主要目的是将状态的行为从主要对象中分离出来,使得主要对象不必包含所有状态的逻辑,而是将每个状态的行为封装在独立的类…...
【校招VIP】产品思维能力之产品设计
考点介绍: 对于产品经理来说最重要的就是产品思维,而拥有一个好的产品思维需要做到以下三点: 1、把握关键点的能力 2、出方案,协调资源,说服团队把资源倾斜到关键点上的能力 3、评估关键点进展程度的能力 『产品思维能…...
微信小程序卡片横向滚动竖图
滚动并不是使用swiper,该方式使用的是scroll-view实现 Swiper局限性太多了,对竖图并不合适 从左往右滚动图片示例 wxml代码: <view class"img-x" style"margin-top: 10px;"><view style"margin: 20rpx;…...
SpringBoot项目(支付宝整合)——springboot整合支付宝沙箱支付 从极简实现到IOC改进
目录 引出git代码仓库准备工作支付宝沙箱api内网穿透 [natapp.cn](https://natapp.cn/#download) springboot整合—极简实现版1.导包配置文件2.controller层代码3.进行支付流程4.支付成功回调 依赖注入的改进1.整体结构2.pom.xml文件依赖3.配置文件4.配置类,依赖注入…...
【AIGC】一款离线版的AI智能换脸工具V2.0分享(支持图片、视频、直播)
随着人工智能技术的爆发,AI不再局限于大语言模型,在图片处理方面也有非常大的进步,其中AI换脸也是大家一直比较感兴趣的,但这个技术的应用一直有很大的争议。 今天给大家分享一个开源你的AI换脸工具2.0,只需要一张所需…...
管理类联考——逻辑——真题篇——按知识分类——汇总篇——一、形式逻辑——选言——相容选言——或
文章目录 第五章 选言命题:相容选言-或;不相容选言-要么要么第一节 选言-相容选言-或-推结论-A或B为真,则非A→B,非B→A(否一则肯一)真题(2010-28)-选言-相容选言-或-推结论-(1)A或B为真,A为假:得B为真(否一则肯一);真题(2012-29)-选言-相容选言-或-推结论-(1)…...
Git如何操作本地分支仓库?
基本使用TortoiseGit 操作本地仓库(分支) 分支的概念 几乎所有的版本控制系统都以某种形式支持分支。 使用分支意味着你可以把你的工作从开发主线上分离开来,避免影响开发主线。多线程开发,可以同时开启多个任务的开发,多个任务之间互不影响。 为何要…...
WPS右键新建没有docx pptx xlsx 修复
解决wps右键没有新建文档的问题 右键没有新建PPT和Excel 1 wps自带的修复直接修复没有用 以上不管咋修复都没用 2 先编辑注册表 找到 HKEY_CLASSES_ROOT CTRLF搜文件扩展名 pptx docx xlsx 新建字符串 三种扩展名都一样操作 注册表编辑之后再次使用wps修复 注册组件&am…...
【巧学C++之西游篇】No.2 --- C++闹天宫,带着“重载“和“引用“
文章目录 前言🌟一、函数重载🌏1.1.函数重载概念🌏1.2.C支持函数重载的原理 -- 名字修饰 🌟二、引用🌏2.1.引用的概念🌏2.2.引用特性🌏2.3.常引用🌏2.4.使用场景🌏2.5.传…...
【HarmonyOS】实现将pcm音频文件进行编码并写入文件(API6 Java)
【关键字】 音频编码、管道模式、createEncoder 【写在前面】 在使用API6开发HarmonyOS应用时,如何将pcm源文件进行编码并写入文件,最后生成aac文件,本文直接附上主要代码开发步骤供大家参考。 【主要功能代码】 import ohos.media.codec.…...
KaiwuDB CTO 魏可伟:回归用户本位,打造“小而全”的数据库
8月16日,KaiwuDB 受邀亮相第十四届中国数据库技术大会 DTCC 2023。KaiwuDB CTO 魏可伟接受大会主办方的采访,双方共同围绕“数据库架构演进、内核引擎设计以及不同技术路线”展开深度探讨。 以下是采访的部分实录 ↓↓↓ 40 多年前,企业的数…...
行业追踪,2023-08-22
自动复盘 2023-08-22 凡所有相,皆是虚妄。若见诸相非相,即见如来。 k 线图是最好的老师,每天持续发布板块的rps排名,追踪板块,板块来开仓,板块去清仓,丢弃自以为是的想法,板块去留让…...
浏览器安装selenium驱动,以Microsoft Edge安装驱动为例
Selenium是一个用于Web应用程序测试的自动化工具。它可以直接在浏览器中运行,模拟真实用户对浏览器进行操作。利用selenium,可以驱动浏览器执行特定的动作,比如:点击、下拉等等,还可以获取浏览器当前呈现的页面的源代码…...
边缘计算网关是如何提高物联网的效率的?
随着物联网的持续发展,物联网应用的丰富和规模的扩大,带来了海量的数据处理、传输和计算需求。 传统的“数据中央处理”模式越来越难以适应物联网的扩展速度,在这一趋势下,边缘计算在物联网系统的部署运营中就发挥出了显著的增效…...
AWVS安装~Windows~激活
目录 1.下载安装包 2.双击acunetix_15.1.221109177.exe进行安装 3.配置C:\Windows\System32\drivers\etc\hosts 4.复制wvsc.exe到C:\Program Files (x86)\Acunetix\15.1.221109177下 5.复制license_info.json与wa_data.dat到C:\ProgramData\Acunetix\shared\license下&…...
ATFX汇市:杰克逊霍尔年会降至,鲍威尔或再发鹰派言论
环球汇市行情摘要—— 昨日,美元指数下跌0.11%,收盘在103.33点, 欧元升值0.22%,收盘价1.0898点; 日元贬值0.58%,收盘价146.23点; 英镑升值0.18%,收盘价1.2757点; 瑞…...
Zipkin开源的分布式链路追踪系统
Zipkin是一款开源的分布式链路追踪系统,主要功能包括: 1. 采集跟踪数据 - Zipkin client库负责收集并上报各服务的请求信息。 2. 存储跟踪数据 - 存储层默认采用Zipkin自带的基于内存的快速存储,也支持整合MySQL、Cassandra等外部存储。 3. 查询接口 - 提供RESTful API进行跟…...
java 项目运行时,后端控制台出现空指针异常---java.lang.NullPointerException
项目场景: 提示:这里简述项目背景: 场景如下: java 项目运行时,后端控制台出现如下图所示报错信息:— 问题描述 提示:这里描述项目中遇到的问题: java 项目运行时,后…...
模型数据处理-数据放入 session和@ModelAttribute 实现 prepare 方法详细讲解
😀前言 本文详细讲解了模型数据处理-数据放入 session和ModelAttribute 实现 prepare 方法详细讲解 🏠个人主页:尘觉主页 🧑个人简介:大家好,我是尘觉,希望我的文章可以帮助到大家,…...
关于android studio 几个简单的问题说明
自信是成功的第一步。——爱迪生 1. android studio 如何运行不同项目是否要更换不同的sdk 和 gradle 2.编译Gradle总是错误为什么 3.如何清理android studio 的缓存 4. 关于android Studio中的build 下面的rebuild project...
angular常用命令
文章目录 1. 创建新项目:2. 生成组件:3. 生成模块:4. 生成服务:5. 运行项目:6. 构建项目:其他 在 Angular 开发中,以下是一些常用的 Angular CLI 命令: 1. 创建新项目: …...
uni-app打包后安卓不显示地图及相关操作详解
新公司最近用uni-app写app,之前的代码有很多问题,正好趁着改bug的时间学习下uni-app。 问题现象: 使用uni-app在浏览器调试的时候,地图是展示的,但是打包完成后,在app端是空白的。咱第一次写appÿ…...
elelementui组件
一、按钮 1、按钮样式 使用type、plain、round和circle属性来定义 Button 的样式。 2、主要代码 <el-row><el-button>默认按钮</el-button><el-button type"primary">主要按钮</el-button><el-button type"success">…...
什么是安全测试报告,怎么获得软件安全检测报告?
安全测试报告 软件安全测试报告:是指测试人员对软件产品的安全缺陷和非法入侵防范能力进行检查和验证的过程,并对软件安全质量进行整体评估,发现软件的缺陷与 bug,为开发人员修复漏洞、提高软件质量奠定坚实的基础。 怎么获得靠谱…...
JS中的Ajax
封装原生 Ajax 请求 在 JavaScript 中,可以通过封装原生的 Ajax 请求来进行与服务器的数据交互。下面是一个封装了原生 Ajax 请求的示例代码,以及对代码的详细注解。 1.简单的Ajax封装代码 <h2>ajax原生</h2><script>//1.创建xhr对象…...
ImportError: cannot import name ‘SQLDatabaseChain‘ from ‘langchain‘解决方案
大家好,我是爱编程的喵喵。双985硕士毕业,现担任全栈工程师一职,热衷于将数据思维应用到工作与生活中。从事机器学习以及相关的前后端开发工作。曾在阿里云、科大讯飞、CCF等比赛获得多次Top名次。现为CSDN博客专家、人工智能领域优质创作者。喜欢通过博客创作的方式对所学的…...
npm、yarn和pnpm
1 node_modules安装方式 在npm3之前是以嵌套结构方式安装依赖包,存在两个问题: 依赖路径太长多个包依赖一个相同包时,本地磁盘会存储多个相同的包 npm3和yarn使用扁平化结构,node_modules变成所有包放在同一层级 注意…...
SparkSQL源码分析系列02-编译环境准备
本文主要描述一些阅读Spark源码环境的准备工作,会涉及到源码编译,插件安装等。 1. 克隆代码。 打开IDEA,在Git下的Clone中,输入 https://github.com/apache/spark,克隆代码到本地,CheckOut到目标版本Spar…...
做网站用什么软件/如何免费搭建自己的网站
这个漏洞比较强,官方说明漏洞只会导致拒绝服务攻击,但实际上利用得当可以实现提权。影响范围3.x-5.x 漏洞成因,数组越界。需要插入用户定义的 index timer set。 XFRM_MSG_NEWSA请求的路劲添加policy。 添加policy需要通过verify_newpolicy…...
注册域名dns/小程序seo推广技巧
先简单解释下什么是双写缓冲。InnoDB使用了一种叫做doublewrite的特殊文件flush技术,在把pages写到date files之前,InnoDB先把它们写到一个叫doublewrite buffer的连续区域内,在写doublewrite buffer完成后,InnoDB才会把pages写到…...
北京网站推广排名公司/为企业推广
越来越多迹象表明,微软将以Microsoft 365 for Consumers捆绑包的形式向消费者提供集成统一的应用程序和服务。据外媒ZDNet的Mary Jo Foley称,该捆绑包包括Office 365个人和家庭产品,不带Windows 10 OS,但带有Teams for Life(Micro…...
求邯郸网站制作/网店培训骗局
好的参考:https://linux.cn/article-4302-1.html http://darkdust.net/files/GDB%20Cheat%20Sheet.pdf 1)加 -g 了吗? 首先请确保在编译时加上了-g 如果出现找不到*.cpp文件等错误,一般是没有正确添加 -g 选项。 2ÿ…...
需要企业网站开发/win7优化
rsh配置&无密码登陆今天才知道我们有一个机器用的是rsh不是ssh(囧),还突然连不上了。就研究了一下,在网上插了好多文章,撞墙好多次终于成功了。现在发布研究结论。步骤如下:1. 安装rsh:#yum…...
bootstrap 3 wordpress theme/线上培训机构排名前十
参考: https://jingyan.baidu.com/article/90bc8fc84b1bb3f653640c05.html...