已配置Anaconda环境，PyCharm中仍无法导入PyTorch怎么办？

Hey there! I’ve run into this exact problem before—Anaconda has PyTorch installed, but PyCharm just won’t recognize it. Let’s walk through the most common fixes step by step:

1. Double-Check PyCharm is Using the Correct Anaconda Environment #

This is the #1 culprit. Here’s how to verify and fix it:

• Open PyCharm, go to File > Settings > Project: [Your Project Name] > Python Interpreter (on Mac, it’s PyCharm > Settings > ...)
• Look at the dropdown at the top—this should be the Anaconda environment where you installed PyTorch (e.g., base or a custom env you created)
• If it’s not the right one:
  • Click the gear icon > Add
  • Select Conda Environment > Existing environment
  • Browse to your Anaconda installation folder, then navigate to envs/[Your Env Name]/python.exe (Windows) or envs/[Your Env Name]/bin/python (Mac/Linux)
  • Hit OK and apply the settings

2. Confirm PyTorch is Actually Installed in Your Anaconda Env #

Let’s make sure the environment has PyTorch first:

• Open Anaconda Prompt (Windows) or your terminal (Mac/Linux)
• Activate your environment with: conda activate [Your Env Name]
• Run conda list torch—you should see a line with torch and its version number
• If it’s missing, reinstall PyTorch in this env (use the command matching your system, e.g., for CPU-only: conda install pytorch torchvision torchaudio cpuonly -c pytorch)

3. Refresh PyCharm’s Package Cache #

PyCharm sometimes doesn’t auto-detect new packages. Here’s how to force a refresh:

• Go back to the Python Interpreter settings page
• Click the circular refresh icon in the top-right corner of the package list
• Wait for PyCharm to scan the environment again—you should see torch pop up in the package list once it’s done

4. Ensure No Project-Specific Virtual Environment is Interfering #

When creating a new project, PyCharm might default to making a fresh virtual environment instead of using your Anaconda env:

• If you already created the project, follow step 1 to switch to the correct Anaconda environment
• For future projects, when setting up, select Previously configured interpreter and pick your Anaconda env from the list

5. Restart PyCharm #

Sometimes the simplest fix works! Close PyCharm completely, reopen it, and try running import torch again. More often than not, this clears up any lingering cache issues.

Give these steps a try—they should get PyCharm and your Anaconda PyTorch environment talking again!

内容的提问来源于stack exchange，提问作者Midoril