pandas Applymap

Python is an excellent language for performing data analysis, mainly due to the strong ecosystem of Python programs focused on handling data. One such tool, pandas, greatly simplifies importing and analyzing data. Each element in a DataFrame is applied with a function using the applymap() method, which receives and returns a scalar. The primary benefit of pandas is the ability to transform data and apply analytics. These functions (map(), applymap(), and apply()) are all used to change the data, but there are some distinctions between them and subtle variations in how they are used. However, we will only discuss the applymap().

How To Use the applymap Function in Windows

The single function allows element-wise action throughout the entire DataFrame is applymap(), which is only accessible in DataFrame.

Although some scenarios perform more quickly than apply() thanks to optimization, comparing the two is still a good idea before undertaking any heavier operations.




func: A single value is returned by a single value in a Python function.

Output: Transformed DataFrame.

Only the DataFrame defines this technique. The dataFrame.applymap() accepts only callable. For DataFrames, applymap() is element-wise. Applymap() is more efficient than Apply(). One piece/element at a time can be operated using the applymap() function.

Example 1: Add Value to Each Element of the Data Frame Using applymap()

Let’s first import an online dataset (iris_data), so we can apply our applymap() function to show the results.

You must import the pandas and add the object “pd” to the code to support the Pandas. Simple mathematical procedures are comparatively easier to perform. According to the axis you define, the function argument when apply() is used on a DataFrame becomes the entire row or column. What if you wanted to apply a specific function on each data frame component rather than just each row and column? When this happens, applymap() is helpful. Consider a situation where a mistake has been committed, and you want to correct it by adding one to each element in your data since you have discovered that the error was a repeatable measurement inaccuracy. To begin, let’s define a helper function:

We have defined an add_one function and passed the item as an argument. It will return item + 1, which means it will add one to each data entry in the dataset. Let’s apply it to every element of the original data frame, excluding the target column, using applymap().

If you contrast this output with the original, you will see that each entry has had one added. It demonstrates applymap() function potential.

Example 2: Calculating the Length of Each Element of the Data Frame Using applymap()

Just like the previous example, we will import pandas and a dataset to start work. Now, what does the applymap() function use? Is it operating on data frames, or will it apply a function to every single element of the data frame? For instance, with this data frame, we created it here. We have all these different values. The applymap() function will take some function and element-wise run it on the single value in this whole data frame and then return a data frame with whatever the result of that function is. Now, since data frames often have data of different types across different columns, the applymap() is a bit less common of a function to use than the apply() or map().

As seen above, we have some numbers in columns and some character strings. So, if we wanted to pass a function that only works on numbers into applymap(), that’s going to throw an error because it can apply to the price column. However, other columns are string. So, it wouldn’t work on that. We have to make sure to pass a function that’s going to work on both numbers and strings. Below, we have defined a function that should work on either numbers or strings:

The previous script will convert these numbers into a string using the str() function. Then, it will just check the length using the len() function. So, we can define this function. If we pass that to applymap(), we will get a data frame as an output having the length of all these different entries in our data frame.

Example 3: Change/Replace the Values of the Data Frame Elements Using the applymap() Function

The pandas’s applymap() function uses a user-defined function to update the data frame’s element-by-element content after receiving a pandas data frame as input. With a dictionary, we can create a lambda function that will give a new value for each element in the data frame as an output to update the values of a data frame. Let’s apply the same illustration to the replace() and map() functions of the pandas library to replace the values in a data frame with those from a dictionary.

By using the sample() function in the random module, we will produce some sample data.

Now, a data frame with a string datatype will be created.

We want to replace some of the data frame’s values with new ones. Here, we will build a dictionary with the new values serving as dictionary values and the old values serving as keys.

We may now update values element-by-element using the pandas applymap() function. The lambda function will be passed as input in the Applymap() function. The lambda function’s input is an element, and its output is the outcome of using a dictionary to query the key.

And as a result, we receive a data frame with updated values.

Example 4: Alter Values/Items of a Data Frame Using the applymap() Function

We’ll first make a data frame with specific rows and columns and specify the names of its indexes.

Let’s multiply each item of the data frame by 10.

Every component of the df DataFrame is multiplied, and the output is saved in the alt df DataFrame. The applymap() function receives a lambda function as a parameter and returns a result by multiplying each item or value by 10. As a result, the DataFrame df’s elements are all scaled by 10. When a nameless function is needed for a brief duration, lambda functions are used. Typically, we utilize it as an argument to a higher-order function in Python. Along with built-in methods like the applymap() and filter(), lambda functions are employed.

We can apply various operations to the DataFrame’s components in addition to mathematical ones.

As you can see in the previous image, we have appended the .99 to every value inside the df data frame. We can also use string values to append the data instead of numeric data.


In this article, we taught how you can use the applymap() function in Python using the pandas library. We have seen that the applymap() method operates on the entire data frame on an element-by-element basis. This taught us how to use the Python pandas dataframe.applymap() method. We worked through examples using this method on the DataFrame to understand how to add values, calculate the length, and replace and alter values of each element of our data frame using the applymap() function.

About the author

Aqsa Yasin

I am a self-motivated information technology professional with a passion for writing. I am a technical writer and love to write for all Linux flavors and Windows.