Hopefully I have dropped this query in the correct place. I am having trouble importing a module which I have already installed. In this case, i used Pip to install Pandas, but when i tried to import Pandas, I receive an error saying that Pandas cannot be imported.
I have checked the version of Python is correct and that my Pip version is correct, i.e. both the latest. I have typed in the command “Pip List” and verified that Pandas 2.2.3 is installed in the packages, but the import function does not work. I have also noticed that the import function does not work for other packages, not just Pandas. So I’m really not sure what i am doing incorrectly, as i have searched Youtube videos, stack overflow and I am not finding anything that will help with this error.
Forgot to mention, i have also visited the Pandas website and checked the dependencies and confirmed i have all dependent packages installed also.
Appreciate any help in resolving or tips.
Are you sure you used the same pip used for the Python interpreter you’re running? If you happen to have more than one Python installed it can get confusing as to what Python interpreter a pip command works with.
What command did you run to install Pandas and what command are you using to run Python? And did you do the install into a virtual environment? And what’s the exact failure you’re seeing from Python not finding Pandas?
Brett is likely correct about your problem. This is why I never run pip, I always run /path/to/python -m pip instead. That way I’m certain that I’m running pip in the same version of Python I’m using.
This mismatch between versions of pip and the interpreter comes up so often that it would be handy if pip could detect if it wasn’t invoked as python -m pip ..., then warn the user about the recommended invocation sequence. I realize detecting that might be cumbersome given the behavior of python -m, but it might still be worth baking into pip.
Maybe not as cumbersome as I thought. When running pip, this might be enough to raise a red flag:
shutil.which('python') == sys.executable
for relevant platform-specific spelling of python. If they aren’t equal, the user could be warned and pip exited. If the user wants to override the result of the test, pip could grow a --force or --interpreter-mismatch-ok flag.
That wouldn’t work on a default Windows install, where python isn’t on PATH and you use py -m pip. And there’s also the pip.pyz zipapp, which is another way you could be invoking pip.
I don’t think there’s a simple answer here, unfortunately. We considered removing the pip command altogether, but experienced users aren’t willing to give up the convenience, even though it’s a source of confusion for beginners 
Note that I waved my hands with “… for relevant platform-specific spelling of python.” 
I don’t know what pip.pyz is, but I trust sys.executable will be available at runtime. It should be relatively straightforward to locate the first python, py, python.exe, … on PATH and compare its location (or the underlying Python executable it runs) with sys.executable.
(Aside–more likely a topic for a private message: Is typing “python” so difficult on Windows that it needs to be shortened to “py”? Said another way, what’s so special about “py” that it’s even required?)
Thanks for your reply Brett!
I think so, but don’t know for sure, see below all of the information I can give you, I can’t seem to see a command that shows the Pip version I used specifically for Pandas, unless you know of one? See below details;
Pip --version
pip 24.3.1 from C:\Users\Prospero\AppData\Local\Programs\Python\Python313\Lib\site-packages\pip (python 3.13)
Python --version
Python 3.13.1
I didn’t install into a virtual environment, I just Command Prompt or Powershell with the Command, “Pip install Pandas”, i used the Ctrl R function to view my command history (not sure if there is a limit to how far the history goes). As for the Python installation, i did this from the Python.org web page. Initially I used the automatic installation for Windows, but I realized this did not install in the correct place and I then modified the installation and selected “Environment Variables”.
As for the error message I am receiving, please see below;
Import pandas as pd
Import : The term ‘Import’ is not recognized as the name of a cmdlet, function, script file, or operable program.
Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
At line:1 char:1
- Import pandas as pd
-
+ CategoryInfo : ObjectNotFound: (Import:String) [], CommandNotFoundException + FullyQualifiedErrorId : CommandNotFoundException
Ah, the issue is you’re trying to type Python code in PowerShell. If you want to do interactive work with Python you need to launch Python’s REPL first. So try typing python3 in PowerShell so you see something like:
Python 3.13.1 (main, Dec 3 2024, 17:59:52) [GCC 11.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>>
From there you can type import pandas at the >>> prompt and see if that works.
Hi Brett,
Thanks so much for your help, I don’t have the issue resolve, but I am making progress. I have now installed the Python interpreter from the Python3 command you gave me.
I have tried to import Pandas from here using the command “Import pandas as pd” and also tried “Import pandas”, I am now receiving the below error;
Import pandas as pd
File “”, line 1
Import pandas as pd
^^^^^^
SyntaxError: invalid syntax
I really appreciate your help!
If you just installed Python then Pandas is probably not installed for the new install.
It’s because you capitalized “Import”; it’s lowercase for the syntax; import pandas (copy and paste that if you want).
Thanks Brett!
for Pandas, do you think it may be easier for me to uninstall and reinstall?
Honestly I think you should use a virtual environment and install into that so you don’t install into your global Python installation. You can either do that manually or use a tool like uv, hatch, or PDM to handle this for you.
Thanks for your help & advice Brett!! Really appreciate it 