Python ModuleNotFoundError Reference — 198+ “No module named X” Fixes (2026)

itsourcecode.com hosts 198+ documented fixes for Python ModuleNotFoundError messages — the most common error every Python developer hits, from beginners running their first script to ML engineers wrestling with TensorFlow imports. Whether you see ModuleNotFoundError: No module named 'numpy', 'tensorflow', 'flask', or some obscure dependency, you’ll find the exact fix below organized by ecosystem. Each post shows the error, the root cause, and 2–4 alternative fixes.

What is ModuleNotFoundError?

A ModuleNotFoundError happens when Python’s import statement can’t find a module on the current Python interpreter’s path. It’s a subclass of ImportError, introduced in Python 3.6. The error message always tells you which module is missing in quotes — for example ModuleNotFoundError: No module named 'pandas'. The fix depends on why Python can’t find it: not installed, installed in a different Python, wrong virtualenv active, typo in the import, or a sub-module that doesn’t exist in the package version you have.

How to fix any ModuleNotFoundError in 5 steps

1. Read the exact module name in quotes. The error tells you what’s missing. Note the spelling — capitalization matters in Python (PIL vs pil, OpenCV vs cv2).
2. Confirm which Python you’re running. Run which python (Mac/Linux) or where python (Windows). Then python --version. If you have multiple Pythons, you may be importing from a different one than where the module is installed.
3. Try installing with pip. Run pip install . If that fails, try python -m pip install — this guarantees you’re using the pip that matches your active Python.
4. Check your virtualenv. If you’re using a virtualenv (venv, conda, poetry), make sure it’s activated before running pip install AND before running your script. Forgotten activation is the #1 cause of this error.
5. Verify the import name matches the package name. Some packages have different install and import names. Example: install pip install beautifulsoup4, but import as from bs4 import BeautifulSoup. Install pip install opencv-python, import as import cv2. Install pip install python-dotenv, import as from dotenv import load_dotenv.

Most common causes (in order of frequency)

• Module not installed — easy fix: pip install
• Wrong Python interpreter active — IDE running system Python but module installed in venv (or vice versa)
• Virtualenv not activated — module installed globally but venv is active, or installed in venv but venv isn’t active
• Install name ≠ import name — see the table above (bs4, cv2, dotenv, etc.)
• Sub-module removed in newer version — common with sklearn, keras, numpy as their internal APIs evolve
• Python core module disabled at build time — _ssl, _tkinter, _curses errors usually mean Python was compiled without those modules
• Python 2 → 3 migration — old codebases importing modules with renamed locations

Featured ModuleNotFoundError fixes by ecosystem

📊 Data Science (numpy, pandas, matplotlib, scikit-learn)

The Python data stack. Most errors here come from sub-module removals between major versions or pip/conda installation mismatches.

• No module named ‘matplotlib’
• No module named ‘sklearn.externals.six’
• No module named ‘numpy.testing.nosetester’
• No module named ‘pyspark.pandas’
• No module named ‘tabulate’
• No module named ‘prettytable’

🤖 Machine Learning & AI (TensorFlow, PyTorch, Keras, OpenCV, LangChain)

ML packages have the most install pain because of GPU/CPU variants, CUDA versions, and rapidly changing internal APIs.

• No module named ‘tensorflow’
• No module named ‘torchvision’ (PyTorch)
• No module named ‘keras.objectives’
• No module named ‘PIL’ (Pillow)
• No module named ‘langchain’
• No module named ‘einops’ — 2026 Guide
• No module named ‘groundingdino’ — 2026 Guide
• No module named ‘pinecone’
• No module named ‘taming’

🌐 Web Frameworks (Flask, Django, FastAPI)

Framework imports usually fail because the framework isn’t installed in the active venv, or you’re importing a sub-package that’s a separate pip install (Flask-CORS, flask_sqlalchemy, etc.).

• No module named ‘flask’
• No module named ‘flask_sqlalchemy’
• No module named ‘flask_cors’ — 2026 Guide
• No module named ‘fastapi’ — 2026 Guide
• No module named ‘aiohttp’
• No module named ‘django_filters’
• No module named ‘debug_toolbar’ (Django)

🗄️ Databases (PostgreSQL, MongoDB, MySQL)

Database connectors often need system libraries before they can be installed. psycopg2 needs libpq-dev on Linux; mysqlclient needs mysql-dev.

• No module named ‘psycopg2’ (PostgreSQL)
• No module named ‘pymongo’ (MongoDB)
• No module named ‘MySQL’ (Python 3)
• No module named ‘mysql’
• No module named ‘flask_sqlalchemy’

🖥️ GUI & Game (Tkinter, PyQt, Pygame)

Desktop GUI errors. _tkinter errors usually mean Python was compiled without Tk support — need to rebuild Python or use a different distribution.

• No module named ‘_tkinter’ — 2026 Guide
• No module named ‘PyQt5.QtWebKitWidgets’
• No module named ‘pygame’

🕷️ Web Scraping (BeautifulSoup, Selenium, Scrapy)

Web scraping has more “install name ≠ import name” gotchas than any other ecosystem. bs4 needs beautifulsoup4; webdriver_manager is its own package.

• No module named ‘bs4’ (BeautifulSoup)
• No module named ‘selenium’
• No module named ‘lxml’
• No module named ‘webdriver_manager’ — 2026 Guide

📦 pip, setuptools, Python core modules

The most painful errors — when Python itself can’t import its own tools. Usually means a broken Python install, missing system dependency, or virtualenv that’s gone bad.

• No module named ‘pip’ in Python
• No module named ‘pip’ on Windows
• No module named ‘pkg_resources’ — 2026 Guide
• No module named ‘setuptools.command.build’
• No module named ‘_ssl’
• No module named ‘_curses’ — 2026 Guide
• No module named ‘distutils.cmd’
• No module named ‘fcntl’ — 2026 Guide

🛠️ Utilities (dotenv, dateutil, click, docx)

Common helper packages. Most of these have classic install/import name confusion.

• No module named ‘dotenv’ (install: python-dotenv)
• No module named ‘python-dateutil’
• No module named ‘click’ — 2026 Guide
• No module named ‘crypto’
• No module named ‘docx’ — 2026 Guide
• No module named ‘ruamel’
• No module named ‘jwt’ — 2026 Guide

🤖 Bots, Notebooks, & Misc

• No module named ‘discord’ — 2026 Guide
• No module named ‘IPython’ — 2026 Guide
• No module named ‘google.colab’ — 2026 Guide

2026 Updated Guides — featured ModuleNotFoundError fixes

These guides were rewritten or expanded in 2026 with current package versions and step-by-step install instructions for Windows, Mac, and Linux.

• No module named ‘fastapi’ — FastAPI install guide
• No module named ‘flask_cors’ — Flask CORS setup
• No module named ‘click’ — Click CLI library
• No module named ‘docx’ — python-docx vs docx confusion
• No module named ‘discord’ — discord.py setup
• No module named ‘einops’ — tensor manipulation
• No module named ‘groundingdino’ — computer vision
• No module named ‘google.colab’ — Colab vs local Python
• No module named ‘IPython’ — Jupyter & notebooks
• No module named ‘jwt’ — PyJWT setup
• No module named ‘fcntl’ — Unix-only on Windows
• No module named ‘pkg_resources’ — setuptools broken
• No module named ‘_tkinter’ — Python compile flags
• No module named ‘_curses’ — terminal libraries
• No module named ‘webdriver_manager’ — Selenium driver auto-management
• No module named ‘debug_toolbar’ — Django toolbar

Frequently asked questions

What’s the difference between ModuleNotFoundError and ImportError?

ModuleNotFoundError is a subclass of ImportError, introduced in Python 3.6. ModuleNotFoundError is raised specifically when the named module can’t be found. ImportError is the broader parent class, also raised when the module IS found but the specific name you tried to import from it doesn’t exist (e.g., from numpy import nonexistent_thing). If you catch ImportError, you’ll also catch ModuleNotFoundError automatically.

Why is pip installing the module but Python still says “No module named X”?

You have multiple Python interpreters on your system, and pip is installing to a different one than the one running your script. Check: (1) Run which pip and which python — do their bin paths match? (2) Use python -m pip install instead of bare pip install — this forces pip to use the same Python as the command. (3) Inside a virtualenv, always activate it BEFORE running pip.

Why does pip install bs4 / opencv / dotenv import differently?

Some Python packages have different install names and import names because the original PyPI name was taken or the package was renamed. Common ones:

• install beautifulsoup4 → import bs4
• install opencv-python → import cv2
• install python-dotenv → import dotenv
• install python-docx → import docx
• install Pillow → import PIL
• install PyYAML → import yaml
• install PyJWT → import jwt

Why am I getting ModuleNotFoundError in VS Code but not in terminal?

VS Code uses its own Python interpreter selection that may differ from your shell’s. Click the Python version indicator in the bottom-right status bar of VS Code → select the interpreter that matches the one where your modules are installed (likely your venv’s bin/python or Scripts/python.exe). After switching, reload the window (Cmd/Ctrl+Shift+P → “Reload Window”).

How do I fix ModuleNotFoundError in Jupyter Notebook?

Jupyter has its own kernel selection separate from terminal Python. Two fixes: (1) Install the module with !pip install inside a notebook cell (the ! runs shell). (2) Make sure the notebook kernel points to the right Python. In Jupyter: Kernel → Change kernel. If the venv isn’t listed, register it with python -m ipykernel install --user --name=myenv.

Can I catch and ignore a ModuleNotFoundError?

Yes, with a try/except, and it’s a legitimate pattern for optional dependencies:

try:
    import optional_module
    HAS_OPTIONAL = True
except ModuleNotFoundError:
    HAS_OPTIONAL = False

But for required dependencies, NEVER catch silently — let the error fail loudly so you know the install is broken.

Why does my Python say “No module named ‘_ssl'” / ‘_tkinter’ / ‘_curses’?

These are core Python modules implemented in C, built into the Python binary at compile time. The error means your Python was compiled WITHOUT the required system library (OpenSSL for _ssl, Tk for _tkinter, ncurses for _curses). Fix: install the system library (e.g., brew install tcl-tk, apt install python3-tk) and rebuild Python, OR use a distribution like pyenv with proper build flags, OR install a different Python (Anaconda, official python.org binary).

How often is this ModuleNotFoundError reference updated?

New posts are added weekly as we hit new errors in real projects. Existing posts are revised every 6–12 months to keep up with package version changes (PyTorch 2.x, TensorFlow 2.16, Pandas 2.x, sklearn 1.x). This page was last refreshed in May 2026.

Related error categories

ModuleNotFoundError is one of 10 hubs in our Python & JavaScript error reference cluster — 980+ documented fixes total. If your error isn’t about a missing module, jump to the right hub below:

• TypeError Reference — 220+ Python & JS TypeError fixes (NoneType, not callable, JS undefined)
• AttributeError Reference — 173+ “object has no attribute X” fixes
• ValueError Reference — 100+ library-specific Python ValueError fixes
• ImportError Reference — 67+ “cannot import name X from Y” fixes (different from ModuleNotFoundError — for version conflicts where the package IS installed)
• NameError Reference — 49+ Python “name X is not defined” fixes
• RuntimeError Reference — 49+ PyTorch CUDA, asyncio, Flask runtime errors
• SyntaxError Reference — 48+ Python & JavaScript parsing errors
• ReferenceError Reference — 34+ JavaScript “is not defined” fixes
• HTTP Error Reference — 35+ HTTP status code fixes
• Python Tutorial — beginner-to-intermediate Python lessons

About this ModuleNotFoundError reference

This ModuleNotFoundError reference has been curated since 2015 by PIES Information Technology Solutions in Binalbagan, Negros Occidental, Philippines. Every post in the collection started as a real error encountered by us, our clients, or readers who emailed us with their tracebacks. Used by 12,000+ Python developers monthly across the Philippines, India, the United States, and beyond. If your error isn’t here, send the full traceback to our contact form and we’ll add it.