astype() for type conversion in Pandas

Introduction

We use astype() to change the type of data in a column or series. This helps us work with data correctly, like turning numbers stored as text into real numbers.

When you have numbers stored as text and want to do math with them.

When you want to convert a column of dates stored as strings into date objects.

When you need to save memory by converting data to a smaller type, like from float64 to float32.

When you want to convert categorical text data into category type for faster processing.

When you want to ensure data types match before merging or analyzing data.

Syntax

Pandas

DataFrame['column_name'] = DataFrame['column_name'].astype(new_type)

new_type can be a Python type like int, float, str, or a NumPy type like np.int32.

You can convert multiple columns by applying astype() on the whole DataFrame with a dictionary of column names and types.

Examples

Convert the 'age' column to integers.

Pandas

df['age'] = df['age'].astype(int)

Convert the 'price' column to floats for calculations.

Pandas

df['price'] = df['price'].astype(float)

Convert multiple columns at once: 'age' to int and 'name' to string.

Pandas

df = df.astype({'age': 'int', 'name': 'str'})

Note: For dates, use pd.to_datetime() instead of astype().

Pandas

df['date'] = pd.to_datetime(df['date'])

Sample Program

This code creates a DataFrame with 'age' and 'height' as strings. Then it converts 'age' to integers and 'height' to floats. Finally, it prints the data types before and after conversion and shows the DataFrame.

Pandas

import pandas as pd

data = {'age': ['25', '30', '22'], 'height': ['5.5', '6.0', '5.8']}
df = pd.DataFrame(data)

print('Before type conversion:')
print(df.dtypes)

# Convert 'age' to int and 'height' to float
 df['age'] = df['age'].astype(int)
 df['height'] = df['height'].astype(float)

print('\nAfter type conversion:')
print(df.dtypes)

print('\nDataFrame content:')
print(df)

Important Notes

If conversion fails (like converting 'abc' to int), astype() will raise an error.

Use errors='ignore' or errors='raise' in some pandas functions, but astype() does not have this parameter.

For converting to category type, use astype('category') to save memory and speed up operations.

Summary

astype() changes the data type of columns or series in pandas.

It helps prepare data for analysis by ensuring correct types.

Always check data before converting to avoid errors.

Practice

1. What does the astype() method do in pandas?

easy

A. It filters rows based on a condition.

B. It changes the data type of a DataFrame column or Series.

C. It sorts the DataFrame by column values.

D. It merges two DataFrames together.

Solution

1. Step 1: Understand the purpose of astype()

   The astype() method is used to convert the data type of a pandas Series or DataFrame column.

2. Step 2: Compare with other options

   Sorting, filtering, and merging are different pandas operations unrelated to astype().

3. Final Answer:

   It changes the data type of a DataFrame column or Series. -> Option B

4. Quick Check:

   astype() changes types = C [OK]

Hint: Remember: astype() changes data types only [OK]

Common Mistakes:

• Confusing astype() with sorting or filtering methods
• Thinking astype() modifies data values instead of types
• Assuming astype() merges DataFrames

2. Which of the following is the correct syntax to convert a pandas Series col to integer type?

easy

A. col.convert(int)

B. astype(col, int)

C. col.astype(int)

D. col.to_int()

Solution

1. Step 1: Recall the method syntax

   The correct syntax to convert a Series is Series.astype(type), so col.astype(int) is correct.

2. Step 2: Check other options

   Options B, C, and D are not valid pandas syntax for type conversion.

3. Final Answer:

   col.astype(int) -> Option C

4. Quick Check:

   Use Series.astype(type) = A [OK]

Hint: Use series.astype(type) to convert types [OK]

Common Mistakes:

• Using astype as a standalone function
• Trying non-existent methods like to_int()
• Passing arguments in wrong order

3. What will be the output of this code?

import pandas as pd
df = pd.DataFrame({'A': ['1', '2', '3']})
df['A'] = df['A'].astype(int)
print(df.dtypes['A'])

medium

A. int64

B. string

C. float64

D. object

Solution

1. Step 1: Understand the initial data type

   Column 'A' starts as strings (object type) because values are in quotes.

2. Step 2: Apply astype(int)

   Converting with astype(int) changes the type to integer, which pandas shows as int64.

3. Final Answer:

   int64 -> Option A

4. Quick Check:

   astype(int) converts to int64 = D [OK]

Hint: astype(int) converts string numbers to int64 type [OK]

Common Mistakes:

• Expecting type to remain object after conversion
• Confusing int64 with float64
• Thinking astype changes values but not type

4. What is wrong with this code snippet?

import pandas as pd
df = pd.DataFrame({'B': ['10', '20', 'thirty']})
df['B'] = df['B'].astype(int)

medium

A. ValueError due to non-numeric string 'thirty'

B. No error, code runs fine

C. SyntaxError because of missing parentheses

D. TypeError because astype() cannot convert strings

Solution

1. Step 1: Identify data in column 'B'

   Column 'B' contains strings '10', '20', and 'thirty'. 'thirty' is not numeric.

2. Step 2: Understand astype(int) behavior

   Trying to convert 'thirty' to int causes a ValueError because it cannot be parsed as a number.

3. Final Answer:

   ValueError due to non-numeric string 'thirty' -> Option A

4. Quick Check:

   Non-numeric string causes ValueError = B [OK]

Hint: Non-numeric strings cause ValueError on astype(int) [OK]

Common Mistakes:

• Assuming astype silently converts all strings
• Confusing ValueError with SyntaxError
• Thinking astype() can convert any string

5. You have a DataFrame df with a column 'score' containing float values as strings like '85.5'. You want to convert this column to numeric floats but also handle any non-numeric values by replacing them with 0.0. Which code correctly does this?

hard

A. df['score'] = df['score'].convert(float, fill=0.0)

B. df['score'] = df['score'].astype(float).fillna(0.0)

C. df['score'] = df['score'].astype(int).fillna(0)

D. df['score'] = pd.to_numeric(df['score'], errors='coerce').fillna(0.0)

Solution

1. Step 1: Understand the problem with astype()

   Using astype(float) fails if there are non-numeric strings, causing errors.

2. Step 2: Use pd.to_numeric() with error handling

   pd.to_numeric() with errors='coerce' converts invalid values to NaN, which can be replaced by fillna(0.0).

3. Step 3: Check other options

   df['score'] = df['score'].astype(float).fillna(0.0) does not handle errors properly; df['score'] = df['score'].astype(int).fillna(0) converts to int, losing decimals; df['score'] = df['score'].convert(float, fill=0.0) uses a non-existent method.

4. Final Answer:

   df['score'] = pd.to_numeric(df['score'], errors='coerce').fillna(0.0) -> Option D

5. Quick Check:

   Use pd.to_numeric with errors='coerce' + fillna = A [OK]

Hint: Use pd.to_numeric with errors='coerce' to handle bad data [OK]

Common Mistakes:

• Using astype(float) without error handling
• Trying to convert to int losing decimals
• Using non-existent methods like convert()