Project Introduction#
A specialized extension of ChatGPT for scientific research work, specifically optimized for enhancing the experience of proofreading academic papers. It supports custom shortcut buttons, custom function plugins, markdown code block table display, dual display of Tex formulas, and new features such as Python/C++ project analysis/translation. It also supports PDF/LaTex paper translation/summarization functions and is compatible with local models such as ChatGLM. Although the original project is already very detailed, it can still be difficult to configure for those who are not familiar with Python. Today, we will provide a detailed guide on configuring this environment using vscode+anaconda.
Project Features#
| Feature | Description |
|---|---|
| One-click Proofreading | Supports one-click proofreading and finding grammar errors in papers |
| One-click Translation between Chinese and English | Supports one-click translation between Chinese and English |
| One-click Code Explanation | Can correctly display and explain code |
| Custom Shortcut Keys | Supports custom shortcut keys |
| Proxy Server Configuration | Supports proxy server configuration |
| Modular Design | Supports custom high-level experimental features and function plugins, with plugin support for hot updates |
| Self-program Profiling | [Function Plugin] One-click understanding of the source code of this project |
| Program Profiling | [Function Plugin] One-click profiling of other Python/C/C++/Java projects |
| Read Papers | [Function Plugin] One-click interpretation of the full text of LaTeX papers and generation of abstracts |
| Batch Comment Generation | [Function Plugin] One-click batch generation of function comments |
| Chat Analysis Report Generation | [Function Plugin] Automatically generates summary reports after running |
| arXiv Assistant | [Function Plugin] Enter the arXiv article URL to translate the abstract and download the PDF with one click |
| PDF Paper Translation | [Function Plugin] Extracts the title and abstract of a PDF paper and translates the full text (multi-threaded) |
| Google Scholar Integration Assistant (Version>=2.45) | [Function Plugin] Given any Google Scholar search page URL, let GPT help you select interesting articles |
| Formula Display | Can display formulas in both Tex format and rendered format |
| Image Display | Can display images in markdown |
| Multi-threaded Function Plugin Support | Supports multi-threaded calling of ChatGPT for one-click processing of massive text or programs |
| Supports Markdown Tables Output by GPT | Can output GPT-supported markdown tables |
| Start Dark Gradio Theme | Add /?__dark-theme=true to the end of the browser URL to switch to the dark theme |
| Huggingface Free Internet Access Online Experience | Log in to Huggingface and copy this space |
| ... | ... |
VSCode+Anaconda Configuration#
1. Clone the project to your local machine#
1. Use Git#
Create a new folder locally and then clone the project directly using a Git tool.
2. Download directly#
2. Configure APIKEY and Proxy#
Here, you need the OpenAI API key and a method for accessing the internet. The original documentation provides a comprehensive explanation.
Tutorial: Complete Beginner's Guide in the Project
3. Configure the Environment in VSCode#
1. Open Anaconda Prompt#
Execute the following commands
-
conda create -n gptac_venv python=3.11
-
conda activate gptac_venv
-
python -m pip install -r requirements.txt
Note: Use the official pip source or Aliyun pip source. Other pip sources (such as some university pip sources) may cause problems. Temporary source switching method:
python -m pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple/
2. Open VSCode#
-
Press ctrl + shift + p to open settings
-
Search for Python interpreter
-
Select the created conda environment
-
Press ctrl + ~ to bring up the terminal and enter python main.py
4. Congratulations!#
Now you can use it happily.
Here are several useful features, such as English academic proofreading, Chinese academic proofreading, Chinese-English translation, and grammar error checking.
PS: The previous ChatGPT account had a free quota of $18 for API usage, but this free quota expired on April 1st. So if your account was registered before, it may show insufficient quota. In this case, there are two solutions:
- Add money to the original account, which can be quite troublesome.
- Create a new account (recommended), which has the lowest cost. The newly created account comes with a $5 quota, which is actually quite a lot and can be used for a long time.