Miniconda is a lightweight version of Anaconda that helps you quickly install Python and various packages, making it more suitable for beginners to run AI programs compared to the full Anaconda distribution.

This tutorial will guide you step-by-step on how to install Miniconda on Windows 10, using the official download link, configure a Python 3.10 environment, and install some commonly used modules. Don't worry, even if you have no prior experience, you can easily get it done!

---

Step 1: Download and Install Miniconda

1. Download Miniconda

   • Open your browser, type in this address, and press Enter: https://www.anaconda.com/download/success#miniconda

   • Scroll down to the "Miniconda" section, click the download link for Windows, or use this direct link: https://repo.anaconda.com/miniconda/Miniconda3-latest-Windows-x86_64.exe

   • The file will be saved to your "Downloads" folder (e.g., C:\Users\YourUsername\Downloads), with a filename similar to Miniconda3-latest-Windows-x86_64.exe.

2. Install Miniconda

   • Double-click the downloaded file to open the installation window.

   • Click "Next" to proceed and agree to the license agreement ("I Agree").

   • Important: On the "Advanced Options" page, check the box "Add Miniconda3 to my PATH environment variable". This allows you to use the conda command directly, otherwise, you might encounter the error conda is not recognized as an internal or external command, operable program or batch file.

   • Note: Some computers may warn against adding to PATH, but ignore it and check the box. This is a common pitfall for beginners; if you don't check it, the commands won't work later.

• Click "Install" and wait for a few minutes. Once completed, click "Finish".

---

Step 2: Verify Installation and Create a Python 3.10 Virtual Environment

1. Open Command Prompt (CMD)

   • Press and hold the Windows key + R on your keyboard to open the "Run" window.
   • Type cmd and press Enter to open the black command-line window.

2. Verify Miniconda Installation

   • In the command line, type: conda --version
   • Press Enter. If it displays something like conda 25.1.1, the installation was successful.

If it says "not recognized as an internal or external command," reinstall and check the Add Miniconda3 to my PATH environment variable box. 3. Create a Python 3.10 Virtual Environment

• In the command line, type: conda create -n myai python=3.10
• myai is the name of the environment, you can choose any name you like (e.g., ai_env).
• After pressing Enter, the system will download Python 3.10 and some basic packages. During the process, it will ask "Y/N". Type y and press Enter to continue the installation.

4. Activate the Virtual Environment
   • Type: conda activate myai
   • Press Enter. If (myai) appears at the beginning of the command line, you have successfully entered the environment.

5. Common Virtual Environment Commands
   • Exit Environment: Type conda deactivate and press Enter. The (myai) will disappear, indicating you've exited.
   • List All Environments: Type conda env list to see a list of your created environments.
   • Delete Environment (if you don't need it): Type conda env remove -n myai to delete it.

---

Step 3: Install Modules using pip and requirements.txt

1. Ensure You Are in the Virtual Environment

   • Type conda activate myai and confirm that the command line shows (myai).

2. Install a Common Module to Test

   • For example, install numpy (a mathematical computation tool required by many AI software packages): pip install numpy Press Enter. After downloading and installing, you can check the version using python -c "import numpy; print(numpy.__version__)".

3. Install from requirements.txt

   • If your AI software provides a requirements.txt file (listing the required packages), copy it to a folder (e.g., a new folder named "AIProject" on your desktop).
   • Navigate to the folder, delete the content in the address bar and type cmd then press Enter to open a command line window in this directory:

• Then type: pip install -r requirements.txt Press Enter, and it will automatically install all the packages listed in the file.

---

Step 4: Common pip Commands and Troubleshooting

1. Common pip Commands

   • Check pip Version: pip --version to see if it's the latest version.
   • Update pip: If the version is old, type pip install --upgrade pip to update.
   • List Installed Packages: pip list to see what's installed.
   • Uninstall a Package: For example, pip uninstall numpy to remove it if you don't need it anymore.

2. Common Errors and Solutions

   • Network Errors (Slow or Failed Downloads):

     • It might be a network issue. Try again a few times, or use a domestic mirror to speed up the download: pip install -r requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple

   • Permission Errors (Access Denied):

     • Run CMD as administrator: Right-click on the "Start" menu, select "Command Prompt (Admin)," and try the command again.

   • Dependency Conflicts (Incompatible Versions):

     • If you get a conflict error for a specific package, try updating pip (pip install --upgrade pip) and then reinstalling. If it still doesn't work, ask the software author for the recommended version.

   • Command Not Recognized:

     • If pip doesn't work, try python -m ensurepip and python -m pip install --upgrade pip to fix it.

---

Now Miniconda is installed, the Python 3.10 environment is configured, and you can install various required packages. If your AI software has running instructions (e.g., python run.py), type the corresponding command in the virtual environment (with (myai) showing on the command line) and give it a try.