Course Recommendations
1. The latest GPT+Python NIR spectroscopy data analysis advanced training course
2. The detailed course content of the ChatGPT in-depth scientific research application, data analysis and machine learning, AI drawing and efficient paper writing training course is as follows:
The latest full-process GPT+Python near-infrared spectroscopy data analysis advanced training course
November 30, 2022, which may become a day that will change human history, is that OpenAI, an American artificial intelligence development agency, launched the chatbot ChatGPT3.5, pushing the development of artificial intelligence to a new level. In April 2023, a stronger version of ChatGPT4.0 will be launched, and multimodal interaction methods such as text, voice, and images will present more possibilities for its application in various industries. On November 7, 2023, OpenAI's first developer conference was called the "Spring Festival Gala of the technology industry", attracting the attention of users around the world, and the GPT store showed OpenAI's ambition to build an AI ecosystem. Therefore, in order to help researchers in the field of near-infrared spectroscopy analysis to be more proficient in the powerful functions of ChatGPT4.0 in near-infrared spectral data analysis, qualitative/quantitative analysis model code automatic generation, etc., and at the same time to learn the basic theoretical knowledge of artificial intelligence (including traditional machine learning, deep learning, etc.) more systematically, as well as specific code implementation methods, an advanced training course on "Near-infrared spectral data analysis supported by the latest AI" is held. It aims to help students master the various methods and skills of ChatGPT4.0 in scientific research, as well as the basic principles and implementation methods of Python and Pytorch codes of classical machine learning algorithms in the field of artificial intelligence (BP neural network, support vector machine, decision tree, random forest, variable dimension reduction and feature selection, group optimization algorithm, etc.) and popular deep learning methods (convolutional neural network, transfer learning, autoencoder, U-Net, etc.). This training adopts a combination of "theoretical explanation + case practice + hands-on practice + discussion and interaction" to explain the latest functions of ChatGPT4.0 in simple terms, as well as the experience and skills that classical artificial intelligence methods need to master in near-infrared spectroscopic data analysis and qualitative/quantitative modeling.
1. Training time and method
On-site time: May 29-June 2, 2024 On-site location: Nanjing
Live broadcast time: May 30 - June 2, 2024 (4 full days of classes)
Live broadcast and on-site training are carried out simultaneously (you can freely choose the mode of participation)
2. Conference welfare
Get a ChatGPT4.0 account that can be directly logged in in China for 1 month [no magic required].
3. Course outline
一、ChatGPT4入门基础
1、ChatGPT概述（GPT-1、GPT-2、GPT-3、GPT-3.5、GPT-4模型的演变）
2. ChatGPT conversation first experience (registration, recharge, purchase method)
3. The difference between GPT-4 and GPT-3.5, and the difference between GPT-4 and domestic large language models (Wenxin Yiyan, Xinghuo, etc.).
4、ChatGPT科研必备插件（Data Interpreter、Wolfram、WebPilot、MixerBox Scholar、ScholarAI、Show Me、AskYourPDF等）
5. Customize your own exclusive GPTs (two ways to make exclusive GPTs: chat/configure parameters, use Knowledge to upload a local knowledge base to improve the performance of exclusive GPTs, use Actions to obtain external information through APIs, and share exclusive GPTs)
6、GPT Store简介
7. Case demonstration and practical exercises
2. ChatGPT4 prompt words use methods and skills
1. ChatGPT Prompt (set the identity for ChatGPT, clarify the task content, provide the background related to the task, give a reference example, specify the format of the returned answer, etc.)
2. Commonly used ChatGPT prompt word templates
3. Template-based ChatGPT prompt word optimization
4、利用ChatGPT4 及插件优化提示词
5. Optimize prompt words through promptperfect.jina.ai
6、利用ChatGPT4 及插件生成提示词
7. ChatGPT4 breaks through the Token limit to receive or output 10,000-word long texts (What is Token?) Conversion between the number of tokens and the number of characters, five ways to submit text that exceeds the token limit, and four ways to make ChatGPT's output break the token limit)
8. Control the output length of ChatGPT (use modifiers, limit the scope of answers, limit the number by context, etc.)
9. Use ChatGPT4 and plug-ins to save your favorite ChatGPT prompt words and call them with one click
10. Case Study: ChatGPT4 is used to realize the design, automatic code generation and operation of the web version of the game
11. Practical exercises
3. ChatGPT4 helps information retrieval and summary analysis
1、传统信息检索方法与技巧总结（Google Scholar、ResearchGate、Sci-Hub、GitHub、关键词检索+同行检索、文献订阅）
2. Use ChatGPT4 and plug-ins to search literature online
3. Use ChatGPT4 and plug-ins to summarize and analyze the content of the literature (three-sentence summary, bullet summary summary, QA summary, table summary, keyword and key sentence extraction, page positioning, multi-document comparison, sentiment analysis)
4. Use ChatGPT4 and plug-ins to summarize Youtube video content
5. Case demonstration and practical exercises
四、ChatGPT4助力论文写作与投稿
1. The overall framework of using ChatGPT4 to automatically generate papers
2. Use ChatGPT4 to complete the translation of the paper (specify the translation role and the field of translation, and give some background hints)
3. Use ChatGPT4 to achieve grammar correction of papers
4. Use ChatGPT4 to complete paragraph structure and sentence logic polishing
5. Use ChatGPT4 to complete the writing and reply to the paper review comments
6. Case demonstration and practical exercises
Five
ChatGPT4助力Python入门基础
1. Python environment construction (download, installation and version selection).
2、如何选择Python编辑器？（IDLE、Notepad++、PyCharm、Jupyter… ）
3. Python Basics (Data Types and Variables, Strings and Encoding, List and Tuple, Conditional Judgment, Loops, Definition and Invocation of Functions, etc.)
4. Installation and use of third-party modules
5. Numpy module library (installation of Numpy; ndarray type properties and the creation of arrays; array indexes and slices; Introduction and use of common functions of Numpy)
6. Matplotlib basic graphic drawing (line chart, bar chart, pie chart, bubble chart, histogram, box chart, scatter chart, etc.), graphic layout (multiple subplot drawing, regular and irregular layout drawing, adding coordinate axes to any position in the canvas)
7. Practical exercises
6. ChatGPT4 helps pre-processing near-infrared spectroscopy data
1. Standardization and normalization of near-infrared spectroscopy data (why do we need standardization and normalization?) ）
2. Handling of outliers and missing values of near-infrared spectral data
3. Discretization and coding of near-infrared spectral data
4. The first and second derivatives of near-infrared spectroscopy data
5. Near-infrared spectral data denoising