I am trying to debug a "ModuleNotFoundError" even though I’ve already installed the package via pip. I suspect Python is looking in the wrong directory or I have multiple versions installed. Is there a simple script or a terminal command that can point me to the exact folder where a specific library is located? I'm on Windows but would also like to know if the method changes for Linux or macOS users.
3 answers
The most straightforward way to find a library's path from within a script is to use the __file__ attribute. For example, if you want to find where pandas is installed, you can run import pandas; print(pandas.__file__). This will return the absolute path to the initialization file of that package. If you prefer using the terminal, pip show <package_name> is incredibly useful as it lists the "Location" along with the version and dependencies. To see the entire list of directories Python searches when you call an import statement, you can run import sys; print(sys.path). This list is the source of truth for your current environment's search priority.
Does the path returned by sys.path include directories added via the PYTHONPATH environment variable, and can we modify this list dynamically during runtime?
On Linux and Mac, you can use the which python or whereis python command in the terminal to find the executable path, which is usually right next to the library folders.
I agree with Nancy, identifying the executable is the first step. For Kimberly's original question, I’ve found that pip show is the most user-friendly way. It gives you the "site-packages" path directly without having to open a Python shell. I always check that location first when I'm dealing with conflicting library versions in my Data Science projects.
Steven, yes it does! sys.path is initialized from the current directory, the PYTHONPATH variable, and the installation-dependent defaults. You can actually modify it at runtime using sys.path.append('/your/custom/path'). This is a common trick in Software Development to import modules from non-standard locations, but use it sparingly as it can make your code harder to debug and less portable across different systems.