Note: This article uses New Bing (GPT4.0) to demonstrate
1. Let him play a Java software developer
Step one: Let ChatGPT play the role of a Java software developer
- Prompt word plug-in: Address: ChatGPT BingChat GPT3 Prompt Generator App (Streamlit) - a Hugging Face Space by Kaludi
Java Software Developer Java Software Developer
I want you to act as a Java software developer. I will provide you with a list of commands and you will implement them. My first request is "I need help creating a Java application."
2. Prepare a Very good Java code example
Step 2: We prepare a poorly written Java code example
- Bad Java Example:
public int calculateUsingIfElse(int a, int b, String operator) { int result = 0; if (operator.equals("add")) { result = a + b; } else if (operator.equals("subtract")) { result = a - b; } else if (operator.equals("multiply")) { result = a * b; } else if (operator.equals("divide")) { if (b == 0) { throw new IllegalArgumentException("Cannot divide by zero"); } result = a / b; } else if (operator.equals("modulus")) { if (b == 0) { throw new IllegalArgumentException("Cannot divide by zero"); } result = a % b; } else if (operator.equals("power")) { result = (int) Math.pow(a, b); } else if (operator.equals("square root")) { if (a < 0) { throw new IllegalArgumentException("Cannot take square root of negative number"); } result = (int) Math.sqrt(a); } else if (operator.equals("absolute value")) { result = Math.abs(a); } else if (operator.equals("negate")) { result = -a; } else { throw new IllegalArgumentException("Invalid operator"); } return result; }
3. Let him optimize this code
Step 3: Let newBing (ChatGPT) optimize this code
- And requires "organizing it into a more elegant code structure, such as automatically splitting sub-functions, reducing if and loop nesting, and even implementing some design patterns."
// 定义一个接口,表示不同的运算策略 public interface OperationStrategy { int apply(int a, int b); } // 定义一个枚举,表示不同的运算符和对应的策略 public enum Operator { ADD((a, b) -> a + b), SUBTRACT((a, b) -> a - b), MULTIPLY((a, b) -> a * b), DIVIDE((a, b) -> { if (b == 0) { throw new ArithmeticException("Cannot divide by zero"); } return a / b; }), MODULUS((a, b) -> { if (b == 0) { throw new ArithmeticException("Cannot divide by zero"); } return a % b; }), POWER((a, b) -> (int) Math.pow(a, b)), SQUARE_ROOT((a, b) -> { if (a < 0) { throw new ArithmeticException("Cannot take square root of negative number"); } return (int) Math.sqrt(a); }), ABSOLUTE_VALUE(Math::abs), NEGATE(a -> -a); private final OperationStrategy strategy; Operator(OperationStrategy strategy) { this.strategy = strategy; } public int apply(int a, int b) { return strategy.apply(a, b); } } // 定义一个方法,根据运算符和操作数来计算结果 public int calculateUsingStrategy(int a, int b, String operatorName) { // 根据运算符名称找到对应的枚举常量,如果没有找到则抛出异常 Operator operator = Operator.valueOf(operatorName.toUpperCase()); if (operator == null) { throw new IllegalArgumentException("Invalid operator"); } // 调用枚举常量的apply方法来执行相应的策略 return operator.apply(a,b); }
4. Let him write a single test
The fourth step: We finally ask him to finish - write a single test Test it
import org.junit.Test; import static org.junit.Assert.*; // 定义一个测试类,用@Test注解标记要执行的测试方法 public class CalculatorTest { // 定义一个测试方法,用assertEquals方法来验证计算结果是否正确 @Test public void testAdd() { int result = calculateUsingStrategy(3, 5, "add"); assertEquals(8, result); } // 定义其他的测试方法,分别测试不同的运算符和操作数 @Test public void testSubtract() { int result = calculateUsingStrategy(10, 4, "subtract"); assertEquals(6, result); } @Test public void testMultiply() { int result = calculateUsingStrategy(7, 9, "multiply"); assertEquals(63, result); } @Test public void testDivide() { int result = calculateUsingStrategy(12, 3, "divide"); assertEquals(4, result); } // 定义一个测试方法,用assertThrows方法来验证抛出了预期的异常 @Test public void testDivideByZero() { assertThrows(ArithmeticException.class, () -> calculateUsingStrategy(10,0,"divide")); } // 定义其他的测试方法,分别测试不同的异常情况 @Test public void testModulusByZero() { assertThrows(ArithmeticException.class, () -> calculateUsingStrategy(10,0,"modulus")); } @Test public void testSquareRootOfNegativeNumber() { assertThrows(ArithmeticException.class, () -> calculateUsingStrategy(-4,-2,"square root")); } // 定义一个测试方法,用assertThrows方法来验证抛出了预期的异常 @Test public void testInvalidOperator() { assertThrows(IllegalArgumentException.class, () -> calculateUsingStrategy(10,-2,"foo")); } }
Finally, data security is the biggest issue. Do not post data to ChatGPT indiscriminately, especially when it involves back-end core storage account passwords, company core business data, and departments. Core strategic planning, etc. Because first of all, ChatGPT will use your question and answer corpus for training; secondly, you can't predict what prompt word the LLM model will answer with the information you inadvertently leaked.
Despite its flaws, the LLM model represented by ChatGPT can greatly improve our development efficiency when acting as our omniscient teacher and tireless universal Util code writer, especially In the fields of data analysis, front-end, unit testing, reconstruction and other fields.
Just like what was written in the first step of the article, ChatGPT is like a versatile identity. You can let him play any role, and Every role can be played in this role range to help us achieve a better life.
More interesting usages are waiting for everyone to discover.
The above is the detailed content of How to make ChatGPT act as a meticulous Java code optimizer?. For more information, please follow other related articles on the PHP Chinese website!

1 前言在发布DALL·E的15个月后,OpenAI在今年春天带了续作DALL·E 2,以其更加惊艳的效果和丰富的可玩性迅速占领了各大AI社区的头条。近年来,随着生成对抗网络(GAN)、变分自编码器(VAE)、扩散模型(Diffusion models)的出现,深度学习已向世人展现其强大的图像生成能力;加上GPT-3、BERT等NLP模型的成功,人类正逐步打破文本和图像的信息界限。在DALL·E 2中,只需输入简单的文本(prompt),它就可以生成多张1024*1024的高清图像。这些图像甚至

Wav2vec 2.0 [1],HuBERT [2] 和 WavLM [3] 等语音预训练模型,通过在多达上万小时的无标注语音数据(如 Libri-light )上的自监督学习,显著提升了自动语音识别(Automatic Speech Recognition, ASR),语音合成(Text-to-speech, TTS)和语音转换(Voice Conversation,VC)等语音下游任务的性能。然而这些模型都没有公开的中文版本,不便于应用在中文语音研究场景。 WenetSpeech [4] 是

“Making large models smaller”这是很多语言模型研究人员的学术追求,针对大模型昂贵的环境和训练成本,陈丹琦在智源大会青源学术年会上做了题为“Making large models smaller”的特邀报告。报告中重点提及了基于记忆增强的TRIME算法和基于粗细粒度联合剪枝和逐层蒸馏的CofiPruning算法。前者能够在不改变模型结构的基础上兼顾语言模型困惑度和检索速度方面的优势;而后者可以在保证下游任务准确度的同时实现更快的处理速度,具有更小的模型结构。陈丹琦 普

由于复杂的注意力机制和模型设计,大多数现有的视觉 Transformer(ViT)在现实的工业部署场景中不能像卷积神经网络(CNN)那样高效地执行。这就带来了一个问题:视觉神经网络能否像 CNN 一样快速推断并像 ViT 一样强大?近期一些工作试图设计 CNN-Transformer 混合架构来解决这个问题,但这些工作的整体性能远不能令人满意。基于此,来自字节跳动的研究者提出了一种能在现实工业场景中有效部署的下一代视觉 Transformer——Next-ViT。从延迟 / 准确性权衡的角度看,

3月27号,Stability AI的创始人兼首席执行官Emad Mostaque在一条推文中宣布,Stable Diffusion XL 现已可用于公开测试。以下是一些事项:“XL”不是这个新的AI模型的官方名称。一旦发布稳定性AI公司的官方公告,名称将会更改。与先前版本相比,图像质量有所提高与先前版本相比,图像生成速度大大加快。示例图像让我们看看新旧AI模型在结果上的差异。Prompt: Luxury sports car with aerodynamic curves, shot in a

人工智能就是一个「拼财力」的行业,如果没有高性能计算设备,别说开发基础模型,就连微调模型都做不到。但如果只靠拼硬件,单靠当前计算性能的发展速度,迟早有一天无法满足日益膨胀的需求,所以还需要配套的软件来协调统筹计算能力,这时候就需要用到「智能计算」技术。最近,来自之江实验室、中国工程院、国防科技大学、浙江大学等多达十二个国内外研究机构共同发表了一篇论文,首次对智能计算领域进行了全面的调研,涵盖了理论基础、智能与计算的技术融合、重要应用、挑战和未来前景。论文链接:https://spj.scien

译者 | 李睿审校 | 孙淑娟近年来, Transformer 机器学习模型已经成为深度学习和深度神经网络技术进步的主要亮点之一。它主要用于自然语言处理中的高级应用。谷歌正在使用它来增强其搜索引擎结果。OpenAI 使用 Transformer 创建了著名的 GPT-2和 GPT-3模型。自从2017年首次亮相以来,Transformer 架构不断发展并扩展到多种不同的变体,从语言任务扩展到其他领域。它们已被用于时间序列预测。它们是 DeepMind 的蛋白质结构预测模型 AlphaFold

说起2010年南非世界杯的最大网红,一定非「章鱼保罗」莫属!这只位于德国海洋生物中心的神奇章鱼,不仅成功预测了德国队全部七场比赛的结果,还顺利地选出了最终的总冠军西班牙队。不幸的是,保罗已经永远地离开了我们,但它的「遗产」却在人们预测足球比赛结果的尝试中持续存在。在艾伦图灵研究所(The Alan Turing Institute),随着2022年卡塔尔世界杯的持续进行,三位研究员Nick Barlow、Jack Roberts和Ryan Chan决定用一种AI算法预测今年的冠军归属。预测模型图


Hot AI Tools

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Undress AI Tool
Undress images for free

Clothoff.io
AI clothes remover

AI Hentai Generator
Generate AI Hentai for free.

Hot Article

Hot Tools

Atom editor mac version download
The most popular open source editor

mPDF
mPDF is a PHP library that can generate PDF files from UTF-8 encoded HTML. The original author, Ian Back, wrote mPDF to output PDF files "on the fly" from his website and handle different languages. It is slower than original scripts like HTML2FPDF and produces larger files when using Unicode fonts, but supports CSS styles etc. and has a lot of enhancements. Supports almost all languages, including RTL (Arabic and Hebrew) and CJK (Chinese, Japanese and Korean). Supports nested block-level elements (such as P, DIV),

SublimeText3 Linux new version
SublimeText3 Linux latest version

VSCode Windows 64-bit Download
A free and powerful IDE editor launched by Microsoft

ZendStudio 13.5.1 Mac
Powerful PHP integrated development environment
