pandas-dev/pandas · error · AssertionError

{cls_name} Expected type {cls}, found {type(left)} instead

Error message

{cls_name} Expected type {cls}, found {type(left)} instead

What it means

Raised by _check_isinstance (asserters.py:162-187), the type-guard used at the top of pandas testing assertion functions (assert_index_equal, assert_series_equal, assert_frame_equal, assert_dict_equal, assert_numpy_array_equal, etc.). It raises AssertionError when the 'left' argument is not an instance of the expected class. This is a test-author error, not a data error.

Source

Thrown at pandas/_testing/asserters.py:181

    """
    Helper method for our assert_* methods that ensures that
    the two objects being compared have the right type before
    proceeding with the comparison.

    Parameters
    ----------
    left : The first object being compared.
    right : The second object being compared.
    cls : The class type to check against.

    Raises
    ------
    AssertionError : Either `left` or `right` is not an instance of `cls`.
    """
    cls_name = cls.__name__

    if not isinstance(left, cls):
        raise AssertionError(
            f"{cls_name} Expected type {cls}, found {type(left)} instead"
        )
    if not isinstance(right, cls):
        raise AssertionError(
            f"{cls_name} Expected type {cls}, found {type(right)} instead"
        )


def assert_dict_equal(left: dict, right: dict, compare_keys: bool = True) -> None:
    _check_isinstance(left, right, dict)
    _testing.assert_dict_equal(left, right, compare_keys=compare_keys)


@set_module("pandas.testing")
def assert_index_equal(
    left: Index,
    right: Index,
    exact: bool | str | lib.NoDefault = lib.no_default,

View on GitHub (pinned to 71959b8cb9)

Solutions

1. Wrap the left argument in the expected constructor, e.g. pd.Series(left_list) or pd.Index(left_array).
2. Use the matching assert function for the actual types (assert_frame_equal for DataFrames, assert_series_equal for Series).
3. Inspect the error message: 'found <type>' tells you exactly what left is.

Example fix

# before
assert_series_equal([1, 2, 3], pd.Series([1, 2, 3]))

# after
assert_series_equal(pd.Series([1, 2, 3]), pd.Series([1, 2, 3]))

import pandas as pd
from pandas.testing import assert_series_equal
left = build_left()
if not isinstance(left, pd.Series):
    left = pd.Series(left)
assert_series_equal(left, expected)

import pandas as pd

def is_series(obj: object) -> bool:
    return isinstance(obj, pd.Series)

When it happens

Trigger: Calling assert_series_equal(list_of_values, series) where left is a Python list instead of a pd.Series; calling assert_index_equal on a plain array; passing a DataFrame where a Series is expected. The function is in pandas.testing and pandas._testing.

Common situations: Test helpers that build expected values as lists/dicts and forget to wrap them in the pandas constructor; refactoring a test and changing the object type on one side; comparing a numpy array with assert_numpy_array_equal but passing a list.

Related errors

• {cls_name} Expected type {cls}, found {type(right)} instead
• check_like must be False if check_index is False
• [datetimelike_compat=True] {left._values} is not equal to {r
• {obj} are different {message}
• {left_base!r} is not {right_base!r}

AI-assisted analysis of pandas-dev/pandas@71959b8cb9 (2026-08-07). Data as JSON: /api/errors/448b2b376af40ab4. Report an issue: GitHub.