1. ChatGPT and AIGC
Generative artificial intelligence - AIGC (Artificial Intelligence Generated Content) refers to technical methods based on artificial intelligence such as generative adversarial networks and large-scale pre-training models. Through the learning and identification of existing data, relevant content is generated with appropriate generalization capabilities. Technology.
The core idea of AIGC technology is to use artificial intelligence algorithms to generate content with certain creativity and quality. By training models and learning from large amounts of data, AIGC can generate relevant content based on input conditions or guidance. For example, by entering keywords, descriptions or samples, AIGC can generate matching codes, articles, images, audio, etc.
ChatGPT (full name: Chat Generative Pre-trained Transformer) is a chat robot program developed by OpenAI and released on November 30, 2022. ChatGPT is a natural language processing tool driven by artificial intelligence technology. It can conduct conversations by understanding and learning human language, and can also interact according to the context of the chat. It can truly chat and communicate like humans, and can even complete writing emails and video scripts. , copywriting, translation, coding, writing papers and other tasks.
Other similar products include: Microsoft's New Bing, Google's Bard, Baidu's Wen Xinyiyan, Alibaba's Tongyi Qianwen, etc.
2. Prompt word project
1. Introduction to Prompt Word Project
Prompt word engineering is a technology that makes appropriate prompts for large model applications so that large models can have better generation effects. It is called Prompt Engineering in English.
At present, prompt word engineering is mainly used in two fields: one is an application similar to a large language model such as ChatGPT, and the other is the field of Vincentian graphs.
Now, whether it is OpenAI ChatGPT, Google Bert, Baidu Wenxinyiyan or Alibaba Tongyi Qianwen, the bottom layer uses the concept of large models and uses a large amount of data for unsupervised pre-training learning. The final result is The trained AI is a generalist. This generalist will give an output result based on our input, and has been predicting the probability of the next token appearing, that is, predicting what will happen next based on the above input. Depending on the input, the quality of the generated results may be completely different.
For example, if you tell a large model application to build a unit portal for you, the effect may be average. But if you tell the big model application and let it act as a product manager, how to make a unit portal, its quality will be much better. It can be seen how important the prompt word project is. The key to big models is asking a good question. Ask a good question and you'll get a good result.
2. Common Tips for Prompt
- Character setting
提示：假定你是xxx方面的专家，...
When writing Prompt, we need to set up the role so that the AI model can better understand our problem.
- clear memory
提示：回复此条对话前，请忽略前面所有的对话
Because ChatGPT is based on the chat model, it will remember or track all the chat records you have written before. Therefore, the use of this sentence can make the subsequent result generation independent of the content of your previous chat.
- Provide requirement details
When writing prompts, we need to provide as much detail as possible so that the AI model can better understand our needs.
- step by step
提示：你必须一步一步地解释所有的事情。
It is mentioned here that "everything must be explained step by step", that is, "step by step". This is so important that "step by step" is known in modern engineering circles as the zero-thinking chain. It will make the AI think step by step and get the results in a logically precise and detailed way, not just general information.
- Easy-to-understand interpretation
提示：假设你是一个从事xxx多年的xxx，而我是一个没有基础的小孩子。
This allows ChatGPT to be explained in language that children can understand.
- Let GPT ask questions
Tip: Before you generate a reply, you can ask me any questions you have
lf you have any questions about this, ask before you try to generate content.Ok?
7**. “Let’s think about it”**
The prompt "Let's think about it", the generated text has the ability to reflect and think deeply, which is very helpful for groups who write prose, poetry, and creative writing.
提示1：让我们想一下气候变化对农业的影响
提示2：让我们讨论下人工智能的现状
提示3：让我们谈一下远程工作的好处和坏处
You can also add an open-ended question, statement, or a text that you want the model to continue, a text that builds on it.
This unique prompt helps ChatGPT give answers from different perspectives and angles, making the output more dynamic and informative.
3. ChatGPT account registration and use
1. Account registration process
2. ChatGPT mirror URL summary
4. Use ChatGPT to review and sort out the test basics
- You are a software testing engineer. I will give you a technical point. Please give: introduction, learning points, learning difficulties, practice questions and answers, and interview questions.
- Introduction to the method of equivalence class division
- Key points in learning the equivalence class division method
- Difficulties in learning the equivalence class division method
- Practice questions and answers for equivalence class division
- Interview questions about equivalence class division method
5. Use ChatGPT to understand project database design and consolidate exercises
- You are a senior software development engineer. Please help me design a database for a project.
- The project is an enterprise asset management system. The functional modules include user login, personal information, storage location management, supplier management, and asset management.
- Permissions include super administrator and asset administrator
- The database is an Oracle database, and the statement to create a user table is
- Insert 10 records into the user table
- Given 10 queries
- Give answers to the 10 questions above
- If you are a software testing interviewer, here are 10 interview questions for database Oracle
- Give answers to the 10 questions above