I'm trying to learn how to use VSCode, after installing Jupyter and trying to add pandas. When I installed VSCode along with Node I got a second version of Python, 3.12.6. Within VSCode I installed Jupyter and trying to follow a tutorial tried importing pandas, but got the 'No Module Found' error.

From what I've read trying to understand the problem, I don't think it's because Jupyter doesn't know which version of Python to use as I've set up a .venv in this Python 3.12.6, the same version that had installed pandas. I used pip show pandas to confirm. It's pandas 2.2.3.

So, do I have to install pandas again within Jupyter? I'm doing it within an .ipynb file which shows '.venv (Python 3.12.6)', so as far as I understand I'm using the right kernel.

What am I missing?

I'm trying to learn how to use VSCode, after installing Jupyter and trying to add pandas. When I installed VSCode along with Node I got a second version of Python, 3.12.6. Within VSCode I installed Jupyter and trying to follow a tutorial tried importing pandas, but got the 'No Module Found' error.

From what I've read trying to understand the problem, I don't think it's because Jupyter doesn't know which version of Python to use as I've set up a .venv in this Python 3.12.6, the same version that had installed pandas. I used pip show pandas to confirm. It's pandas 2.2.3.

So, do I have to install pandas again within Jupyter? I'm doing it within an .ipynb file which shows '.venv (Python 3.12.6)', so as far as I understand I'm using the right kernel.

What am I missing?

• python
• visual-studio-code
• jupyter-notebook

Share Improve this question Follow edited Mar 7 at 16:07 wjandrea 33.2k1010 gold badges6969 silver badges9898 bronze badges asked Mar 6 at 17:50 IchoryDickIchoryDick 322 bronze badges 11

• 1 Your environment situation is very complex here and adding in Jupyter which is also particular about the kernel environment adds another layer. (Although I don't think it deserves a new post on a topic well-covered here and so don't be surprised if it gets closed.) To get past the hurdle, it may just be easier to install Pandas in the specific environment where the kernel is using. If you don't have Anaconda/conda in the mix here, try %pip install pandas in the notebook on the cell before you want to import pandas. The magic install commands (there's a %conda one, too) ensure the .... – Wayne Commented Mar 6 at 19:18
• <continued> installation occurs in the environment where the kernel is running and can often be easier to use then connecting your Jupyter to the correct environment where you previously installed a package outside of Jupyter. For more help see here and here. Maybe also see here, too? – Wayne Commented Mar 6 at 20:02
• @Wayne You should post %pip install pandas as an answer :) along with some of the related info like the links – wjandrea Commented Mar 7 at 0:11
• 1 Okay. I can try the %pip install pandas. As I don't fully understand what happens I didn't want to risk perhaps duplicating installations as I see the problems people have in having more than one Python version. To @wjandrea, yes. Pandas was installed first. It is a while ago that I installed VSCode, and Node. I can't remember the sequence now. Pandas was installed at that time, along with the newer version of Python. Setting up the .venv is recent, as I was following a VSCode tutorial and it was recommended. I'd appreciate it if you could post your solutions. – IchoryDick Commented Mar 7 at 12:16
• 1 @IchoryDick "more than one Python version" - Slight correction: you only have one Python version, but two environments that use it (global and virtual). The virtual environment uses the global interpreter as its base. – wjandrea Commented Mar 7 at 16:20

 |  Show 6 more comments

1 Answer 1

Sorted by: Reset to default 0

Since you installed Pandas before setting up the virtual environment, it was installed in the global environment and is not available to the virtual environment. There are three way to solve this:

1. Install Pandas again into the virtual environment.

   This is the best option if you're starting a project that depends on Pandas, because it gives you control over dependency versioning, i.e, if the project relies on features in specific versions of Pandas, those are part of the project and not the OS, where Pandas could get upgraded or even downgraded if you switch OSes.

   • In Jupyter (IPython), you can use %pip install pandas to install within the current kernel (current environment).

     If you were using conda, you would use %conda install pandas

   • Outside Jupyter, you can activate the virtual environment then use pip install pandas (or conda install pandas for a conda env).

     You may want to enable python.terminal.activateEnvironment in VSCode to have it automatically activate the environment in the internal terminal.

2. Use the global environment.

   If you don't really care about the Pandas version and don't want to futz with a virtual environment, this is fine.

3. Let the virtual environment access the global environment.

   This sort of defeats the purpose of a virtual environment, so I would only do this as a last resort.

Related

• Python environments in VS Code - VSCode docs
• Why do I get a "ModuleNotFoundError" in VS Code despite the fact that I already installed the module?
• venv Python stdlib module docs - Gives an overview of virtual environments