Fix pandas Has No Attribute DataFrame: Import and Naming

Quick answer: pandas does expose the DataFrame constructor. When code reports that the module has no attribute DataFrame, inspect what was imported before reinstalling anything: a local pandas.py or pandas directory may shadow the real package, the name may be misspelled, or the program may be using a different interpreter than the one where pandas was installed.

Python Pool infographic troubleshooting pandas DataFrame import local shadowing package version and environment — The correct pandas constructor is pandas.DataFrame; inspect the imported module and local filenames before reinstalling the package.

AttributeError: module 'pandas' has no attribute 'dataframe' almost always means the code used the wrong capitalization. The pandas class is DataFrame with a capital D and capital F. Python attribute names are case-sensitive, so pd.dataframe and pd.DataFrame are different names.

Quick fix

Use pd.DataFrame(), not pd.dataframe().

import pandas as pd

# Wrong
# df = pd.dataframe({"name": ["Ada"], "score": [95]})

# Correct
df = pd.DataFrame({"name": ["Ada"], "score": [95]})
print(df)

The official pandas API documents pandas.DataFrame as the primary two-dimensional tabular data structure. There is no lowercase pandas.dataframe constructor.

Why this error happens

When you write import pandas as pd, pd points to the pandas module. Python then looks for the exact attribute name after the dot. If the code asks for pd.dataframe, pandas cannot find that lowercase attribute and raises the error.

import pandas as pd

print(hasattr(pd, "DataFrame"))   # True
print(hasattr(pd, "dataframe"))   # False

Check for a local pandas.py file

If the capitalization is already correct but the error still looks strange, check whether your project contains a file named pandas.py or a folder named pandas. That local file can shadow the real pandas package.

import pandas as pd
print(pd.__file__)

If the printed path points to your project folder instead of the installed package location, rename the local file, delete any related __pycache__ files, and restart Python.

Python Pool infographic showing pandas module, DataFrame class, lowercase dataframe typo, and AttributeError — The class is named DataFrame with a capital D and F; Python attribute names are case-sensitive.

Make sure pandas is installed in the active environment

Another common cause is running code in a different interpreter, virtual environment, or Jupyter kernel from the one where pandas was installed. Install pandas with the same Python executable that runs your code.

python -m pip install pandas

In conda environments, use:

conda install -c conda-forge pandas

If you are using Jupyter Notebook, restart the kernel after installing pandas. For related environment checks, see our guides on checking the Python version and fixing python is not recognized.

Fast troubleshooting flow

Start with the smallest check first. If changing pd.dataframe to pd.DataFrame fixes the error, you do not need to reinstall pandas. Reinstall only after confirming the name is correct and pandas is imported from the expected location.

Search the file for dataframe and replace only constructor calls with DataFrame.
Run import pandas as pd in the same script, shell, or notebook cell.
Print pd.__file__ to confirm Python imported the real package.
Check python -m pip show pandas in the same environment.
Restart the IDE or Jupyter kernel after changes.

This order avoids a common mistake: reinstalling pandas when the real issue is only a lowercase class name or a local file shadowing the package.

Python Pool infographic comparing import pandas as pd, pd.DataFrame, local module, and constructed table — Use the conventional pandas alias and call the class from that module object.

Jupyter Notebook checklist

Run import sys; print(sys.executable) inside the notebook.
Install pandas with that executable: python -m pip install pandas.
Restart the kernel after installation.
Check that you did not name the notebook or a nearby file pandas.py.

Do not confuse DataFrame with to_numpy()

If you are converting a DataFrame to a NumPy array, use df.to_numpy(). Do not call a DataFrame object like a function, and do not replace the constructor with lowercase dataframe.

import pandas as pd

df = pd.DataFrame({"score": [95, 88]})
array = df.to_numpy()
print(array)

For conversion-focused examples, read our guide on converting a NumPy array to a pandas DataFrame.

Common fixes

Change pd.dataframe(...) to pd.DataFrame(...).
Confirm the import is import pandas as pd.
Rename any local pandas.py file that shadows the real package.
Install pandas in the active environment with python -m pip install pandas.
Restart your terminal, IDE, or Jupyter kernel after changing the environment.

Python Pool infographic showing project file named pandas, import resolution, wrong module, and missing attribute — A local file or folder named pandas can shadow the installed package and produce misleading attributes.

Related pandas and NumPy fixes

Official references

Use The Public Constructor And Check Case

The supported spelling is pandas.DataFrame, commonly written as pd.DataFrame after import pandas as pd. Python attribute lookup is case-sensitive, so Dataframe and dataframe are not equivalent. Keep the import near the top of a small reproduction and remove unrelated notebook state.

import pandas as pd

frame = pd.DataFrame({"value": [1, 2, 3]})
print(frame)
print(type(frame).__name__)

Python Pool infographic testing module path, package version, capitalization, environment, and validation — Check the resolved module path, package version, capitalization, active environment, and a minimal constructor test.

Inspect The Module That Was Imported

Print the module’s file and version in the same process that fails. If __file__ points into your project, a file or directory is shadowing the dependency. Rename the local path, remove stale bytecode if appropriate, and restart the interpreter so its import cache is rebuilt.

import pandas as pd
import sys

print("python:", sys.executable)
print("pandas:", pd.__file__)
print("version:", getattr(pd, "__version__", "unknown"))
print("has DataFrame:", hasattr(pd, "DataFrame"))

Compare Installation And Runtime

An editor, notebook kernel, shell, and production process can use different interpreters. Run the package inspection through the same executable that runs the code. Do not mix a system pip install with a virtual environment and then infer that pandas itself is missing.

import subprocess
import sys

subprocess.run([sys.executable, "-m", "pip", "show", "pandas"], check=False)
print(sys.executable)

Avoid Accidental Name Replacement

A later assignment such as pandas = {} or pd = [] can replace the imported module in a notebook or function scope. Search for assignments to the module name, restart the kernel, and prefer descriptive local names so a dependency cannot be mistaken for a data object. The same issue can come from a function parameter, a loop variable, or a test fixture that reuses the short alias. A clean process and a minimal reproduction make the real import contract visible, while a large notebook can preserve a stale object long after the source cell has been corrected.

import pandas as pd

data = pd.DataFrame({"value": [10, 20]})
# Do not reuse pd or pandas for a list, dictionary, or DataFrame.
print(data["value"].sum())

Use the official pandas.DataFrame documentation as the API reference. When the attribute is absent, environment and name resolution are more useful first checks than changing capitalization at random.

For related pandas failures, compare modern DataFrame combination, boolean-mask validation, and the Python visualizer workflow when isolating an environment or API mismatch.

Frequently Asked Questions

Does pandas have a DataFrame constructor?

Yes. The public constructor is pandas.DataFrame, so this error usually points to a shadowed import, wrong package, or incorrect capitalization.

Why does import pandas as pd still fail?

Check that pd is the real pandas module, that no local pandas.py or pandas directory shadows it, and that the active interpreter has pandas installed.