System information
Mac OS X 10.12 Sierra
Pycharm 2016.2.3
Python 3.5.1
Jupyter 4.1.0
Question
When I'm using Jupyter Notebook under Pycharm, it doesn't seem to include my current working directory, or the source root directory, in the PYTHONPATH of the currently active Jupyter Notebook. This is unlike the behaviour of ordinary .py files, for which Pycharm automatically includes its current directory and the project source root directory in the PYTHONPATH. So, is it possible to set Pycharm to automatically include the project source root directory in the PYTHONPATH for Jupyter Notebooks run under Pycharm?
It turns out, in a past endeavour, I modified my jupyter_notebook_config.py in %USER%\.jupyter folder. Looked at lines as follows:
## The directory to use for notebooks and kernels.
c.NotebookApp.notebook_dir = 'D:\\Works\\GitHub\\Jupyter\\'
The line under the ##, indicates the override path, evidently.
When I commented it out...Pycharm all of the sudden successfully launched with pwd of the file (location of the .ipynb). good luck, hth.
Pycharm 2019.1 has done an overhaul of Jupyter support. Jetbrains says this in reference to Pycharm 2019.1: "Jupyter Notebook support in PyCharm was redesigned from the ground up."
So, as of Pycharm 2019.1, you can now set Pycharm to include the project source root directory for Jupyter Notebooks.
See this page of Pycharm 2019.1 help.
There's a bunch of ways to mess with the environment, but the PyCharm specific way that worked for me was to make the directory with my source in it a PyCharm "sources root" as described here: https://www.jetbrains.com/help/pycharm/content-root.html
This can be done from the context menu of a directory in the navigator, or from the "Project Structure" settings pane.
This seems to update the PYTHONPATH as well as sorting out things like autocomplete.
For anyone using the free version of PyCharm, which has no direct Jupyter Notebook support, there's a simple workaround:
Go to PyCharm "Terminal" tab at the bottom (NOT to be confused with the Python console tab!)
[NOTE: steps 2 and 3 are for Window7; on other OS's, just research how to set environment variables thru the terminal or command prompt]
Issue the command:
set PYTHONPATH=\path_to_your_project_root_folder\
EXAMPLE on Windows: set PYTHONPATH=\Docs\- MY CODE\my_project_root_folder\
Optionally, you may verify that the environmental variable got set by issuing the command: set PYTHONPATH
Issue the command: jupyter notebook
That's all! A new browser tab will open up... and, insider any notebook, you can verify the path by issuing the python commands:
import sys
sys.path
Related
I want Jupyter Lab to use the directory I launch it from on the command prompt. However, it seems to be defaulting to C:\Users\[USERNAME]\AppData\Local\Temp.
I confirm this by running the following:
import inspect
inspect.getfile(inspect.currentframe())
always returns 'C:\\Users\\[USERNAME]\\AppData\\Local\\Temp\\ipykernel_16200\\3653608800.py' regardless of the directory that I launched Jupyter Lab from. Any ideas on how to fix this?
Details:
On other computers/Anaconda installs, I have been able use this command sequence successfully to obtain the working directory and then point to other files relative to it (i.e., add to the system path to support importing modules). I don't understand why it isn't working on this new install. I have not run a --generate-config command, so there does not appear to be any relevant global config settings interfering here.
I am running Windows 10 with Anaconda 2022.05, JupyterLab version 3.3.2
The solutions presented here do not work for me:
Launching with jupyer lab . does not fix the issue
Launching with jupyter lab --notebook-dir=. does not fix the issue
Launching with jupyter lab --notebook-dir=$pwd does not fix the issue
The solution is found in #rv.kvetch's and #Carlo's comments:
Use the os.getcwd() function instead of inspect.currentframe() (or use pwd!).
This code does what I need for finding my working directory (and adding relative paths):
import os
current_folder = os.getcwd()
My Flask App server is running but I have three imports that cannot be resolved.
I have tried:
reinstalling the imports individually
reinstalling requirements.txt
I configured VSCode Workspace with an extra path to my project folder (there is now a .vscode file within the root of my project folder where it was not before)
I have updated my venv path settings in VSCode
Here is my file structure:
- > .vscode
- > client *(React front end)*
- > data
- > server *(Python/Flask back end)*
- > app
- > venv
- config.py
- README.md
- requirements.txt *(this contains the 3 unresolved, along with several that are resolving)*
- .env
- .flaskenv
- .gitignore
- requirements.txt
Unfortunately none of these things have resolved my imports issue and my routes are still not working. Any ideas/suggestions?
Open the Command Palette (Ctrl+Shift+P), then select the Python: Select Interpreter. From the list, select the virtual environment in your project folder that starts with .env.
Run Terminal: Create New Integrated Terminal (Ctrl+Shift+` or from the Command Palette), which creates a terminal and automatically activates the virtual environment by running its activation script.
Install sqlalchemy and mongoengine with command pip install. Once installing them successfully, there will intellisense when you import them and no warnings shown.
Besides, the folder .vscode is to store Workspace settings as well as debugging and task configurations.
If you are using a virtual environment, and even after trying pip installing all the necessary libraries, you have to select the python interpreter that exists in the virtual environment folder.
(Ctrl+Shift+P) then search for "Python: Select Interpreter"
Click "Enter interpreter path" followed by "Find.."
Navigate to your project virtual environment folder
Go into "Scripts" folder and then select "python.exe" as the interpreter.
These steps allow you to select the right python interpreter associated with the project's virtual environment.
Open the Command Palette (Ctrl+Shift+P), then select Python: Clear Cache and Reload window.
Voila, all the import errors vanished.
I specified a path to the python interpreter I'm using within the settings.json file contained in the project repo's .vscode folder.
"python.pythonPath": "path-to-interpreter.python.exe"
Thanks to the following resource! https://dev.to/climentea/how-to-solve-pylance-missing-imports-in-vscode-359b
I ran into this error after an upgrade of my local python version (brew -> manual install), even though the specified interpreter was already /usr/local/bin/python3.
CMD + Shift + P and re-selecting the [same] interpreter fixed the error.
In hindsight, I suspect that a restart of VSCode could have also fixed this. 🤷♂️
Perhaps VSCode is using the incorrect Python path for this reason. A base interpreter should be used instead of the vscode interpreter, if necessary.
After verifying my Python interpreter was sourced correctly in VS Code, I simply cleaned my env and reinstalled the project locally and the import errors disappeared. I personally had an odd VS Code/Pylance cache and reinstalling the dependencies and modules fixed it for me.
I was getting the very same error you have and none of the solutions presented here worked me.
I work on a windows machine. I use miniconda to manage my virtual environments. And when I'm coding I launch every program from the command prompt (cmd), including Vscode.
Even tough inside Vscode the python evironment was correctly settled I was getting the very same import error that you mention. The interesting part of it was that I wasn't having any problems to run the code, it was working as usuall. But when coding I wasn't able to see the functions related to the libraries with the import error.
The solution:
Activate the correct conda env for the project before launch Vscode from the command prompt.
Why does this happen?
I BELIVE that this happen because when you don't activate any conda env before launch Vscode. The base conda env is loaded as default. That generates the import conflict. (You can chek this out installing those packagues into the environment and running everything just as you have been doing)
When I create a new project and the virtual environment using the venv python package, VSCode doesn't recognize the new virtual environment. I follow the bellow instruction:
https://code.visualstudio.com/docs/python/environments
The command that I use in the VSCode integrated terminal is:
python -m venv .venv
The terminal that I use is PowerShell 7, But I tried the CMD terminal too.
After running this command, the .venv folder is created very well on the workspace and I checked its behavior on the terminal.
I tried conda package manager to create a venv and VSCode recognizes it. The problem is only with the venv Python package.
I also tried another Python version to create venv, But the problem still exists.
I read this question:
How can I set up a virtual environment for Python in Visual Studio Code?
I know how to add an environment manually, but I expect VSCode to recognize my environments automatically.
I added these lines to settings.json (Preferences):
"python.venvPath": "~/.venv",
"python.venvFolders": [
"~/.venv/Scripts"
]
I activated the venv manually using VSCode integrated terminal.
The problem still isn't solved.
This is a screenshot of my problem:
I know how to add an environment manually, but I want it to be automatic.
I tried VSCode on another PC, and it worked; It doesn't need anything to do except the presence of venv.
VSCode Version: 1.58.0
Python Extension Version: v2021.6.944021595
The "python.venvPath" should be set to the parent folder of the .venv folder.
Virtual environments located in the folder identified by the
python.venvPath setting (see General settings), which can contain
multiple virtual environments. The extension looks for virtual
environments in the first-level subfolders of venvPath.
But, you need not set that. The Python extension should find the venv environment, as it is just located under your workspace folder.
Virtual environments located directly under the workspace (project)
folder.
I tried it locally, it does not work too. It will only show the cached environment path, and the "python.venvPath" does not work either.
And there are some changes of the interpreter storage:
A VSCode internal storage is introduced which will now store the
interpreter settings in the workspace & workspace folder scope.
You can refer to here for more details.
So, it's recommended to select the environment path manually for now.
After writing this command in VS CodeTerminal
python -m venv env
You can simply use
.\env\Scripts\Activate
In the same directory path where you have created the virtual enviroment.
For some reason VSCode had uninstalled all my Python extensions including the main Python extension. After reinstalling them it was able to detect my virtualenvs again.
So I was having this same issue and I resolved it by,
1. Open Settings in VSCode.
2. Search for "python.terminal.activateEnvironment"
3. If unchecked/unticked, then check/tick for both "User" and "Workspace".
The problem is not with VSCode but the workspace settings in which I was working on.
Just changed that and Voila works like a charm.
I just bought my very first Mac after using Windows my entire life and have been trying to install things. I am also still kind of a beginner at programming.
I installed Anaconda for Python 2.7 version, and then I installed the 3.7 version right after.
I tried going to terminal and typing jupyter notebook but it says:
zsh: command not found: jupyter
I am able to launch a jupyter notebook (I think) when I search for jupyter notebook in spotlight and executing the top hit.
I've been reading a lot on forums about this issue and see a lot of people talking about PATH which I am not really sure about.
I've seen people also post their which python information and theirs seem very different from mine:
which python3
usr/bin/python3
which python
usr/bin/python
Any help would be appreciated! Still learning how to set up this Mac. Thank you.
You might want to refer to this link https://jupyter-notebook-beginner-guide.readthedocs.io/en/latest/execute.html
3.1.2. Change Jupyter Notebook startup folder (Mac OS)
To launch Jupyter Notebook App:
Click on spotlight, type terminal to open a terminal window.
Enter the startup folder by typing cd /some_folder_name.
Type jupyter notebook to launch the Jupyter Notebook App The notebook interface will appear in a new browser window or tab.
From what I can understand you are not changing to the startup directory of Jupyter.
You need to add jupyter to your path.
In your terminal, type in the following command:
export PATH="/PUT JUPYTER PATH HERE:$PATH"
Then try jupyter notebook
Reasoning: If installing using pip install --user, you must add the user-level bin directory to your PATH environment variable in order to launch jupyter lab
I would like to use Jupyter notebook inside Pycharm. The project interpreter is a python2.7 from a virtual environment inside WSL (ubuntu 18.04).
The Jupiter package is correctly installed inside the virtual environment (I can run it by jupyter notebook).
My problem is that when I want to use Jupyter notebook inside Pycharm, I get the following error: Run Error Jupyter package is not installed (see picture).
Any idea what's going on here?
I had this problem in Python 3. Below are the steps I took to resolve the issue; I believe they should resolve the issue for you too:
I had Jupyter Lab installed. Pycharm only works with Jupyter Notebook. Long story short, if you have Jupyter Lab installed you need to uninstall all your packages using:
$ pip freeze | xargs pip uninstall -y
Restart your computer
Follow Jupyter Notebook installation instructions
Make sure WSL is set up through pycharm instructions: wsl pycharm instructions
In Pycharm, open an .ipynb file. Click the dropdown that says "Managed Jupyter server" It's right above the text editor. Select "configure Jupyter server". Check configured server.
In your wsl terminal, type jupyter notebook. Copy and paste the text that looks like: http://localhost:8888/?token=874asdf687asd6fasd8f74ds6f4s9d8f7sddf into the cofigured server box in Pycharm.
That's it. You should be able to run the jupyter cells in pycharm now.
I have Pycharm 2020.3 For me the issue was I was using a virtual environment with "inherit global site packages." I had Jupyter installed in global site packages but NOT the virtual environment.
Once I installed Jupyter within the virtual environment Jupyter notebook worked. Not sure why inheriting Jupyter from global packages wasn't working for me.
The above solution using a designated url with token seems to work with older versions of PyCharm. A simpler solution is to upgrade to the latest PyCharm. I no longer had an issue with the auto server using PyCharm 2019.3.2 (Mac)
I had this problem with Datalore plugin enabled on 2020.2 linux, running on bare metal but displaying to a remote X server (probably doesn't matter). My solution was to disable the Datalore plugin (it's enabled for professional pycharm by default).
This way I was still able to use the "managed" auto-start version with better integration / debugging vs the "configured" option (or at least with less hassle).
Note since it's been a year, my problem is probably different than OP.
This happened for me, when the interpreter was a remote one. I fixed this by changing the interpreter to one from a local env.
This can be done by selecting the Configure Jupyter Server.
I also meet this problem,and i solved it
i create the new project with the global sit-packages like below
then i meet the problem
i create the new project with no global sit-packages and install jupyter notebook in the virtualenv
then the problem is gone