2

I'm trying to load a module called folium into my Jupyter Notebook environment. When I attempt to load this module, a ModuleNotFoundError is returned.

When I run !pip list in the same Jupyter Notebook environment, folium is listed amongst my installed modules.

How do I successfully load this module?

Improve this question
4
  • 3
    As I advised here, try %pip install folium in a cell inside your notebook. Then restart the kernel. If that doesn't help, try the same with %conda install -c conda-forge folium. If you haven't encountered use of % with pip & conda before inside notebooks, see here. Essentially, don't use ! anymore in the notebook with pip & conda. You are better off with nothing ibefore pip or conda in notebook as automagics will handle. Commented Jan 25, 2022 at 18:35
  • To follow up on that, what ! does is send out that task to your system's shell I believe, and so what it may be listing with pip list there isn't necessarily what is in the environment backing the notebook kernel. I think. Or the installation is botched and that is why it can show installed with !pip list and not work. (Would be interesting before you run my suggested solution steps above if you ran !pip list and then %pip list in your notebook and see if they show different things in regards to folium.) Anyway, using the magics is always better now. Commented Jan 25, 2022 at 18:43
  • 1
    Thanks for the advice. When I run %pip list I cannot see folium listed, so clearly folium isn't installed in the environment backing the notebook kernel. I've installed it using %pip install folium, and can now load it successfully. Can you suggest a resource where I can learn about how and where packages are installed (i.e. so that I can learn from first principles why I had this issue). Thanks for all your help. Commented Jan 26, 2022 at 11:37
  • Sorry, I don't have anything that immediately springs to mind for this. In part because there's a lot of ways you can be using environments depending on how you installed things and how you work on your system. Did you install with Anaconda? If not, did you have Jupyter previously installed & recently hit this issue where you are thinking you installed packages and they are unavailable? I'm guessing since you are asking about understanding where packages are installed, you are not using virtual environments yourself directly; however, as you look into environment stuff you may see that option. Commented Jan 26, 2022 at 17:04

1 Answer 1

Reset to default
2

Could it be that you expect a certain Python installation (virtual environment) to be used in Jupyter, but Jupyter is using a different kernel?

With that assumption, I believe all you need to do is to add the virtual environment to Jupyter as a kernel. In more detail:

  • Install & activate your virtual environment. Below I'll use virtualenv, which nowadays seems to be the recommended way of working (see Python docs), but there are other options as well:
    • pip install --user virtualenv if it is not installed yet (it is installed by default for the newer Python versions). By the way, personally, I use Poetry instead of pip for a while now, and I like it a lot.
    • Go to the desired directory (with cd <path>) and create the environment with python3 -m venv <my_venv_name>.
    • Activate the environment with source <my_venv_name>/bin/activate (deactivate with deactivate).
  • Make sure ipykernel is installed (pip install --user ipykernel)
  • Add the venv to Jupyter with python3 -m ipykernel install --user --name=<any_name_referring_to_your_venv>.

Now you should be able to select in Jupyter this environment as a kernel, and all packages installed in that venv should be picked up by Jupyter automatically as well (you'll need to restart the kernel in that case though).

Improve this answer
Sign up to request clarification or add additional context in comments.

Comments

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.