In this post, we will see how to convert column to float in Pandas.
Table of Contents
Convert String column to float in Pandas
There are two ways to convert String column to float in Pandas.
Using asType(float) method
You can use asType(float) to convert string to float in Pandas.
Here is the syntax:
|
1 2 3 |
df['Column'] = df['Column'].astype(float) |
Here is an example. We will convert data type of Column Rating from object to float64
Sample Employee data for this Example.
| Name | Rating |
|---|---|
| Andy | 8.6 |
| Prateek | 7 |
|
1 2 3 4 5 6 7 8 9 10 |
import pandas as pd emp_df = pd.DataFrame({'Name': ['Andy','Prateek'], 'Rating': ['8.6','7']}) print("-------Original Dataframe-------\n",emp_df) print("-------Original Dataframe's data types-------\n",emp_df.dtypes) emp_df['Rating'] = emp_df['Rating'].astype(float) print("-------Updated Dataframe with float data type for Rating column-------\n",emp_df) print("-------Updated Dataframe's data types-------\n",emp_df.dtypes) |
Output:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 |
-------Original Dataframe------- Name Rating 0 Andy 8.6 1 Prateek 7 -------Original Dataframe's data types------- Name object Rating object dtype: object -------Updated Dataframe with float data type for Rating column------- Name Rating 0 Andy 8.6 1 Prateek 7.0 -------Updated Dataframe's data types------- Name object Rating float64 dtype: object |
Using toNumeric() method
You can use toNumeric() method where you might have non-numeric values in the column.
Here is the syntax:
|
1 2 3 |
df['Column'] = pd.to_numeric(df['Column'],errors='coerce') |
If column as non-numeric values, then they will be converted to NaN.
Here is sample Employee Data for this example.
| Name | Rating |
|---|---|
| Andy | 8.6 |
| Riyanka | NA |
| Prateek | 7 |
| Neha | NA |
|
1 2 3 4 5 6 7 8 9 10 |
import pandas as pd emp_df = pd.DataFrame({'Name': ['Andy','Riyanka','Prateek','Sneha'], 'Rating': ['8.6','NA','7','NA']}) print("-------Original Dataframe-------\n",emp_df) print("-------Original Dataframe's data types-------\n",emp_df.dtypes) emp_df['Rating'] = pd.to_numeric(emp_df['Rating'],errors='coerce') print("-------Updated Dataframe with float data type for Rating column-------\n",emp_df) print("-------Updated Dataframe's data types-------\n",emp_df.dtypes) |
Output:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 |
-------Original Dataframe------- Name Rating 0 Andy 8.6 1 Riyanka NA 2 Prateek 7 3 Sneha NA -------Original Dataframe's data types------- Name object Rating object dtype: object -------Updated Dataframe with float data type for Rating column------- Name Rating 0 Andy 8.6 1 Riyanka NaN 2 Prateek 7.0 3 Sneha NaN -------Updated Dataframe's data types------- Name object Rating float64 dtype: object |
As you can see, we have converted Rating column to float64.
Convert Integer column to float in Pandas
There are two ways to convert Integer column to float in Pandas.
Using asType(float) method
You can use asType(float) to convert string to float in Pandas.
Here is the syntax:
|
1 2 3 |
df['Column'] = df['Column'].astype(float) |
Here is an example. We will convert data type of Column Salary from integer to float64
|
1 2 3 4 5 6 7 8 9 10 11 12 |
import pandas as pd emp_df = pd.DataFrame({'Name': ['Rohan','Abhishek'], 'Salary': [80000,40000]}) print("-------Original Dataframe-------\n",emp_df) print("-------Original Dataframe's data types-------\n",emp_df.dtypes) # Converting int to float data type emp_df['Salary'] = emp_df['Salary'].astype(float) print("-------Updated Dataframe with float data type for Salary column-------\n",emp_df) print("-------Updated Dataframe's data types-------\n",emp_df.dtypes) |
Output:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 |
-------Original Dataframe------- Name Salary 0 Rohan 80000 1 Abhishek 40000 -------Original Dataframe's data types------- Name object Salary int64 dtype: object -------Updated Dataframe with float data type for Salary column------- Name Salary 0 Rohan 80000.0 1 Abhishek 40000.0 -------Updated Dataframe's data types------- Name object Salary float64 dtype: object |
Using toNumeric() method
You can use toNumeric() method where you might have non-numeric values in the column.
Here is the syntax:
|
1 2 3 |
df['Column'] = pd.to_numeric(df['Column'], downcast='float') |
Here is an example:
|
1 2 3 4 5 6 7 8 9 10 |
import pandas as pd emp_df = pd.DataFrame({'Name': ['Neha','Amrita'], 'Salary': [70000,60000]}) print("-------Original Dataframe-------\n",emp_df) print("-------Original Dataframe's data types-------\n",emp_df.dtypes) emp_df['Salary'] = pd.to_numeric(emp_df['Salary'], downcast='float') print("-------Updated Dataframe with float data type for Rating column-------\n",emp_df) print("-------Updated Dataframe's data types-------\n",emp_df.dtypes) |
Output:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 |
-------Original Dataframe------- Name Salary 0 Neha 70000 1 Amrita 60000 -------Original Dataframe's data types------- Name object Salary int64 dtype: object -------Updated Dataframe with float data type for Rating column------- Name Salary 0 Neha 70000.0 1 Amrita 60000.0 -------Updated Dataframe's data types------- Name object Salary float32 dtype: object |
That’s all about pandas convert column to float.
