Python NameError Reference — 49+ “name X is not defined” Fixes (2026)

itsourcecode.com hosts 49+ documented fixes for Python NameError messages — the error class raised when Python encounters a name (variable, function, class, or module reference) that isn’t defined in the current scope. Most NameErrors trace to missing imports (np, pd, keras), Python 2 → 3 migration leftovers (raw_input, unicode), or Jupyter-only builtins running outside a notebook (display, get_ipython). Browse the fixes below by category.

What is a Python NameError?

A NameError is raised when Python tries to use a name that hasn’t been defined yet in the local, enclosing, or global scope. The error message always names the missing identifier in quotes — for example NameError: name 'np' is not defined means you wrote np.array(...) but never did import numpy as np. NameError is a sibling of UnboundLocalError (variable defined later in function) and a subclass of Exception.

How to fix any NameError in 4 steps

1. Read the missing name in quotes. The error tells you exactly which identifier is undefined.
2. Check if you imported it. If the name is np, pd, plt, tf, nn, or any library alias — you almost certainly need an import library as alias line at the top of the file.
3. Check spelling and case. Python is case-sensitive. True ≠ true. List ≠ list. A typo IS a NameError.
4. Check the scope. A variable defined inside a function isn’t visible outside it. A name defined in a Jupyter cell that hasn’t run yet doesn’t exist. A name in another module needs to be imported, not just typed.

Featured NameError fixes by category

📊 Library imports (numpy, pandas, matplotlib, sklearn)

The most common NameError — you used a library alias but forgot the import.

• name ‘np’ is not defined (numpy)
• name ‘pd’ is not defined (pandas)
• name ‘keras’ is not defined
• name ‘model’ is not defined
• name ‘spark’ is not defined (PySpark)
• name ‘sqlcontext’ is not defined (PySpark SQL)

🐍 Python built-ins and stdlib

• name ‘os’ is not defined
• name ‘sys’ is not defined
• name ‘time’ is not defined
• name ‘random’ is not defined
• name ‘timedelta’ is not defined (datetime)
• name ‘open’ is not defined
• name ‘list’ is not defined
• name ‘base’ is not defined

📓 Jupyter / IPython-only names

Names that exist inside Jupyter notebooks but raise NameError when you copy the code to a regular .py script.

• name ‘display’ is not defined (IPython display)
• name ‘get_ipython’ is not defined
• name ‘__file__’ is not defined (Jupyter / REPL)

🔁 Python 2 → 3 migration

Names that existed in Python 2 but were removed or renamed in Python 3.

• name ‘raw_input’ is not defined (use input())
• name ‘unicode’ is not defined (Python 3 uses str)

🔣 Scope and self-reference

• name ‘self’ is not defined (method vs function)
• name ‘kwargs’ is not defined
• name ‘_C’ is not defined (PyTorch internal)
• name ‘x_train’ is not defined (ML notebook scope)
• name ‘X’ is not defined

2026 Updated Guides

• Name ‘model’ is not defined — ML notebook scope issues
• Name ‘spark’ is not defined — PySpark session setup
• Name ‘list’ is not defined — built-in shadowing
• Name ‘__file__’ is not defined — Jupyter/REPL context

Related error categories

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

• TypeError Reference — 220+ Python & JS TypeError fixes
• ModuleNotFoundError Reference — 198+ Python import errors
• AttributeError Reference — 173+ “object has no attribute X” fixes
• ValueError Reference — 100+ pandas/NumPy/sklearn fixes
• ImportError Reference — 67+ “cannot import name X from Y” fixes
• RuntimeError Reference — 49+ PyTorch/CUDA/async runtime errors
• SyntaxError Reference — 48+ Python & JavaScript syntax errors
• ReferenceError Reference — 34+ JavaScript “is not defined” fixes
• HTTP Error Reference — 35+ HTTP status code fixes (4xx, 5xx)
• Python Tutorial — beginner-to-intermediate Python lessons

About this NameError reference

This NameError reference has been built since 2015 by PIES Information Technology Solutions in Binalbagan, Negros Occidental, Philippines. Every post comes from a real error encountered in production code. Used by 12,000+ Python developers monthly across the Philippines, India, the United States, and beyond.