TypeError: unhashable type 'slice' in Python [Solved]

# TypeError: unhashable type 'slice' in Python [Solved]

The "TypeError: unhashable type 'slice'" exception in Python occurs for 2 main reasons:

Trying to slice a dictionary, e.g. a_dict[:2].
Trying to slice a DataFrame object, e.g. df[:, 2].

If you got the error when slicing a DataFrame object in pandas, click on the following subheading:

TypeError: unhashable type 'slice' when slicing a DataFrame (pandas)

# Slicing a Dictionary in Python

The syntax for list slicing is a_list[start:stop:step].

main.py

Copied!
a_list = ['bobby', 'hadz', 'com']

print(a_list[:2]) # 👉️ ['bobby', 'hadz']

print(a_list[1:3]) # 👉️ ['hadz', 'com']

The start index is inclusive and the stop index is exclusive (up to, but not including).

The slicing syntax can be used with objects such as lists, tuples and strings, however, it cannot be used with a dictionary.

Here is an example of how the error occurs when trying to slice a dictionary.

main.py

Copied!
a_dict = {
    'id': 1,
    'name': 'bobby hadz',
    'age': 30,
}

# ⛔️ TypeError: unhashable type: 'slice'
print(a_dict[:2])

trying to slice dictionary

One way to get around the error is to manually access the key-value pairs.

main.py

Copied!
a_dict = {
    'id': 1,
    'name': 'bobby hadz',
    'age': 30,
}

employee_id = a_dict['id']
print(employee_id)  # 👉️ 1

employee_name = a_dict['name']
print(employee_name)  # 👉️ bobby hadz

You can access specific key-value pairs using bracket notation.

# Using dict.items() to slice a dictionary

Alternatively, you can use the items() method to slice a dictionary.

main.py

Copied!
a_dict = {
    'id': 1,
    'name': 'bobby hadz',
    'age': 30,
}

a_slice = list(a_dict.items())[:2]

# 👇️ [('id', 1), ('name', 'bobby hadz')]
print(a_slice)

for key, value in a_slice:
    print(key, value)

slice dictionary with item

Notice that we used the list() class to convert the dictionary's items to a list before slicing.

This is necessary because the dict_items object cannot directly be sliced.

The dict.items() method returns a new view of the dictionary's items ((key, value) pairs).

main.py

Copied!
my_dict = {'id': 1,  'name': 'BobbyHadz'}

# 👇️ dict_items([('id', 1), ('name', 'BobbyHadz')])
print(my_dict.items())

If you need to iterate over a slice of the dictionary, use a for loop.

main.py

Copied!
a_dict = {
    'id': 1,
    'name': 'bobby hadz',
    'age': 30,
}

a_slice = list(a_dict.items())[:2]

for key, value in a_slice:
    print(key, value)

The code sample produces the following output.

output

Copied!
id 1
name bobby hadz

# Creating a dictionary containing a slice of the original dict

You can use a for loop if you need to create a dictionary that contains a slice of the original dict.

main.py

Copied!
a_dict = {
    'id': 1,
    'name': 'bobby hadz',
    'age': 30,
}

dict_slice = {}

for key, value in list(a_dict.items())[:2]:
    dict_slice[key] = value

# 👇️ {'id': 1, 'name': 'bobby hadz'}
print(dict_slice)

using dict items method to create slice

The code sample uses a for loop to iterate over the slice and adds each key-value pair to a new dictionary.

Alternatively, you can pass the list slice directly to the dict() class.

main.py

Copied!
a_dict = {
    'id': 1,
    'name': 'bobby hadz',
    'age': 30,
}

a_slice = list(a_dict.items())[:2]
print(a_slice)  # 👉️ [('id', 1), ('name', 'bobby hadz')]

dict_slice = dict(a_slice)
print(dict_slice)  # 👉️ {'id': 1, 'name': 'bobby hadz'}

We used the dict class to convert the list of key-value pairs to a Python dictionary.

# Getting a slice of a dictionary's keys or values

You can use the dict.keys() and dict.values() method to slice the dictionary's keys or values.

main.py

Copied!
a_dict = {
    'id': 1,
    'name': 'bobby hadz',
    'age': 30,
}

keys_slice = list(a_dict.keys())[1:3]
print(keys_slice) # 👉️ ['name', 'age']

values_slice = list(a_dict.values())[1:3]
print(values_slice) # 👉️ ['bobby hadz', 30]

Notice that we used the list() class to convert the dictionary's keys and values to lists before slicing.

This is necessary because the dict_keys and dict_values objects cannot directly be sliced.

The dict.keys() method returns a new view of the dictionary's keys.

main.py

Copied!
my_dict = {'id': 1,  'name': 'BobbyHadz'}

print(my_dict.keys())  # 👉️ dict_keys(['id', 'name'])

The dict.values() method returns a new view of the dictionary's values.

main.py

Copied!
my_dict = {'id': 1,  'name': 'bobbyhadz'}

print(my_dict.values())  # 👉️ dict_values([1, 'bobbyhadz'])

# TypeError: unhashable type 'slice' when slicing a DataFrame

You might also get the "TypeError: unhashable type 'slice'" when slicing a pandas DataFrame object.

main.py

Copied!
import pandas as pd


data = {
    'id': [1, 2, 3, 4],
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'salary': [100, 50, 75, 150],
    'experience': [5, 7, 9, 13],
    'task': ['dev', 'test', 'ship', 'manage'],
}


df = pd.DataFrame(data)

# ⛔️ pandas.errors.InvalidIndexError: (slice(1, 3, None), slice(None, None, None))
print(df[1:3, :])

The df variable stores a DataFrame and can't be sliced directly.

Instead, you should use the iloc attribute to slice the DataFrame.

main.py

Copied!
import pandas as pd


data = {
    'id': [1, 2, 3, 4],
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'salary': [100, 50, 75, 150],
    'experience': [5, 7, 9, 13],
    'task': ['dev', 'test', 'ship', 'manage'],
}


df = pd.DataFrame(data)

print(df.iloc[1:3, :])

The code sample produces the following output.

output

Copied!
   id   name  salary  experience  task
1   2  Bobby      50           7  test
2   3   Carl      75           9  ship

using iloc property to slice dataframe

iloc is used to select rows and columns by integer location.

There is also a loc attribute that is used for label-based indexing.

main.py

Copied!
import pandas as pd


data = {
    'id': [1, 2, 3, 4],
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'salary': [100, 50, 75, 150],
    'experience': [5, 7, 9, 13],
    'task': ['dev', 'test', 'ship', 'manage'],
}


df = pd.DataFrame(data)

print(df.loc[1:3, ['name', 'salary']])

The code sample produces the following output.

output

Copied!
    name  salary
1  Bobby      50
2   Carl      75
3    Dan     150

loc is used to select rows and columns by label(s).

using loc indexer instead

You can also access a DataFrame using bracket notation if you need to print the values of a specific column.

main.py

Copied!
import pandas as pd


data = {
    'id': [1, 2, 3, 4],
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'salary': [100, 50, 75, 150],
    'experience': [5, 7, 9, 13],
    'task': ['dev', 'test', 'ship', 'manage'],
}


df = pd.DataFrame(data)

print(df['name'])

The code sample produces the following output.

output

Copied!
0    Alice
1    Bobby
2     Carl
3      Dan
Name: name, dtype: object

# Slicing a `DataFrame` when encoding categorical data

Here is another example of how the error occurs.

main.py

Copied!
import pandas as pd
from sklearn.preprocessing import LabelEncoder


data = {
    'id': [1, 2, 3, 4],
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'salary': [100, 50, 75, 150],
    'experience': [5, 7, 9, 13],
    'task': ['dev', 'test', 'ship', 'manage'],
}


df = pd.DataFrame(data)

y = df.iloc[:, 4]
x = df.iloc[:, 0:4]

labelencoder_X = LabelEncoder()

# ⛔️ TypeError: unhashable type: 'slice'
x[:, 3] = labelencoder_X.fit_transform(df.iloc[:, 3])

print(df.iloc)

The error is caused by the following line.

main.py

Copied!
x[:, 3] = labelencoder_X.fit_transform(df.iloc[:, 3])

The x variable stores a DataFrame object and they cannot be sliced directly.

To solve the error, use the iloc attribute for position-based indexing.

main.py

Copied!
import pandas as pd
from sklearn.preprocessing import LabelEncoder


data = {
    'id': [1, 2, 3, 4],
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'salary': [100, 50, 75, 150],
    'experience': [5, 7, 9, 13],
    'task': ['dev', 'test', 'ship', 'manage'],
}


df = pd.DataFrame(data)

y = df.iloc[:, 4]
x = df.iloc[:, 0:4]

labelencoder_X = LabelEncoder()

# ✅ Accessing the `iloc` attribute
x.iloc[:, 3] = labelencoder_X.fit_transform(df.iloc[:, 3])

We only changed the following line in the code sample.

main.py

Copied!
x.iloc[:, 3] = labelencoder_X.fit_transform(df.iloc[:, 3])

iloc is used to select rows and columns by integer location.

You can also access the values attribute before slicing.

The values attribute returns the NumPy representation of the DataFrame object.

main.py

Copied!
import pandas as pd
from sklearn.preprocessing import LabelEncoder


data = {
    'id': [1, 2, 3, 4],
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'salary': [100, 50, 75, 150],
    'experience': [5, 7, 9, 13],
    'task': ['dev', 'test', 'ship', 'manage'],
}


df = pd.DataFrame(data)

y = df.iloc[:, 4]
x = df.iloc[:, 0:4]

labelencoder_X = LabelEncoder()

# ✅ Accessing values attribute
x.values[:, 3] = labelencoder_X.fit_transform(df.iloc[:, 3])

We only changed the last line of the code sample.

The values attribute returns a numpy.ndarray object that stores the values of the DataFrame.

The to_numpy method can also be used to convert the DataFrame object to a NumPy array.

main.py

Copied!
import pandas as pd
from sklearn.preprocessing import LabelEncoder


data = {
    'id': [1, 2, 3, 4],
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'salary': [100, 50, 75, 150],
    'experience': [5, 7, 9, 13],
    'task': ['dev', 'test', 'ship', 'manage'],
}


df = pd.DataFrame(data)

y = df.iloc[:, 4]
x = df.iloc[:, 0:4]

labelencoder_X = LabelEncoder()

# ✅ Calling to_numpy()
x.to_numpy()[:, 3] = labelencoder_X.fit_transform(df.iloc[:, 3])

We used the to_numpy() method to convert the DataFrame object to a NumPy array prior to using slicing notation.

Want to learn more about working with dictionaries in Python? Check out these resources: Get the Last Key or Value in a Dictionary in Python,Get multiple values from a Dictionary in Python.

# Conclusion

To solve the "TypeError: unhashable type 'slice'" exception:

Convert the dictionary's items to a list before slicing.
Use the iloc attribute on a DataFrame object before slicing.

# Additional Resources

You can learn more about the related topics by checking out the following tutorials:

I wrote a book in which I share everything I know about how to become a better, more efficient programmer.

You can use the search field on my Home Page to filter through all of my articles.

TypeError: unhashable type 'slice' in Python [Solved]