Filling NA values

As you remember, removing all the rows with NA values is not always the best solution. Let's learn another method, how to fill NA values in a dataframe.

If you want to fill NAs with specific values, use the .fillna() method. There are many options on what parameters you should use. Let's consider obligatory and the most appropriate ones:

value - defines value that should be set instead of NAs. Can be a dictionary in format {'column_name': value_to_replace}.
method - defines the method that should be used to fill NA values (for instance, 'ffill': propagate last valid observation forward to next valid backfill / 'bfill': use next valid observation to fill gap.
limit - number of consequtive NA values that should be filled. By default, this parameter isn't specified. For instance, we can fill the NA values within the 'omphtotinch', 'hmwkswk' columns with respective means.

Everything was clear?

Section 2. Chapter 7

Course Content

Data Manipulation using pandas

1. Preprocessing Data: Part I

What is Data Preprocessing?Types consistency Poor Data Presentation Manipulating Strings Challenge Replacing Specific Elements Simultaneous Replacement Challenge

2. Preprocessing Data: Part II

Logical Inconsistency Removing Rows Challenge Outliers Challenge Missing Values Filling NA values Challenge

3. Grouping Data

What is Grouping Data?Grouping in pandas [1/2]Challenge Grouping in pandas [2/2]Challenge Grouping by Several Columns Challenge

4. Aggregating and Visualizing Data

Advanced Aggregation [1/2]Challenge Advanced Aggregation [2/2]Challenge Histograms Challenge Bar and Scatter Plots Other Types of Graphs Challenge 1 Challenge 2

5. Joining Data

What is Joining Data?Left Join Right Join Inner Join Outer Join Concatenation

Filling NA values

As you remember, removing all the rows with NA values is not always the best solution. Let's learn another method, how to fill NA values in a dataframe.

If you want to fill NAs with specific values, use the .fillna() method. There are many options on what parameters you should use. Let's consider obligatory and the most appropriate ones:

value - defines value that should be set instead of NAs. Can be a dictionary in format {'column_name': value_to_replace}.
method - defines the method that should be used to fill NA values (for instance, 'ffill': propagate last valid observation forward to next valid backfill / 'bfill': use next valid observation to fill gap.
limit - number of consequtive NA values that should be filled. By default, this parameter isn't specified. For instance, we can fill the NA values within the 'omphtotinch', 'hmwkswk' columns with respective means.

Everything was clear?

Section 2. Chapter 7