CSV files are great for data. They can sometimes contain missing values. Pandas provides ways to handle these. This ensures clean data import.
Default Missing Values
Pandas recognizes some values as missing by default. These include empty strings and “NaN”.
Custom Missing Values
You might have other values representing missing data. You can specify these using the na_values argument in pd.read_csv.
Example (Custom Missing Values)
import pandas as pd
data = pd.read_csv("my_data.csv", na_values=["N/A", "missing"]) # Replace with your file path
This code tells Pandas that “N/A” and “missing” should also be treated as missing values.
Keeping Default NaN Values
Sometimes you want to keep the default “NaN” recognition. You can ensure this with the keep_default_na argument. Setting it to True keeps the default behavior.
Example (Keeping Default NaN)
import pandas as pd
data = pd.read_csv("my_data.csv", na_values=["N/A"], keep_default_na=True) # Replace with your file path
This example treats “N/A” as missing. It also keeps the default “NaN” recognition.
For more complex scenarios, check the read_csv documentation. It has more advanced options.