Introduction
Working with information is a giant a part of any information evaluation venture. In Python, the Pandas library is a strong instrument that gives versatile and environment friendly information buildings to make the method of information manipulation and evaluation simpler. Probably the most widespread information buildings supplied by Pandas is the DataFrame, which could be considered a desk of information with rows and columns. Nonetheless, usually you will wish to save your DataFrame to a file for later use, or to share with others. Probably the most widespread file codecs for information storage is CSV.
On this article, we’ll discover learn how to write a pandas DataFrame to a CSV file.
Why Write a DataFrame to a CSV File?
CSV information are a preferred alternative for information storage for quite a lot of causes. At the start, they’re text-based and due to this fact human-readable. This implies you possibly can open a CSV file in a plain textual content editor to rapidly view and perceive the info it comprises.
CSV information are additionally broadly used and understood by many various software program functions. This makes it straightforward to share information between totally different techniques and programming languages. In the event you’re working with a crew that makes use of a wide range of instruments, saving your DataFrame to a CSV file ensures that everybody can work with the info.
Lastly, writing a DataFrame to a CSV file is a technique to persist your information. Whenever you’re working in a Python session, your DataFrame exists solely in reminiscence. In the event you shut your Python session, your DataFrame is misplaced. By writing it to a CSV file, it can save you your information to disk, permitting you to entry it once more later, even after you have closed and reopened your Python session.
import pandas as pd
df = pd.DataFrame({
'A': [1, 2, 3],
'B': ['a', 'b', 'c']
})
df.to_csv('my_data.csv')
On this code, a DataFrame is created after which written to a CSV file named my_data.csv. After working this code, you will discover a new file in your present listing with this title, containing the info out of your DataFrame.
Methods to Write a DataFrame to a CSV File
Pandas, a preferred Python information manipulation library, supplies a easy but highly effective methodology to write down a DataFrame to a CSV file. The perform to_csv() is what we want.
Let’s begin with a primary DataFrame:
import pandas as pd
information = {'Title': ['John', 'Anna', 'Peter'],
'Age': [28, 24, 33],
'Nation': ['USA', 'Sweden', 'Germany']}
df = pd.DataFrame(information)
Our DataFrame appears like this:
Title Age Nation
0 John 28 USA
1 Anna 24 Sweden
2 Peter 33 Germany
To put in writing this DataFrame to a CSV file, we use the to_csv() perform like so:
df.to_csv('information.csv')
This can create a CSV file named information.csv in your present listing.
If you wish to specify a distinct location, present the complete path. For instance, df.to_csv('/path/to/your/listing/information.csv').
Writing DataFrame to CSV with Particular Delimiter
By default, the to_csv() perform makes use of a comma as the sector delimiter. Nonetheless, you possibly can specify a distinct delimiter utilizing the sep parameter.
For instance, let’s write our DataFrame to a CSV file utilizing a semicolon because the delimiter:
df.to_csv('data_semicolon.csv', sep=';')
This can create a CSV file named data_semicolon.csv with the info separated by semicolons.
Title;Age;Nation
John;28;USA
Anna;24;Sweden
Peter;33;Germany
Notice: The sep parameter accepts any character as a delimiter. Nonetheless, widespread delimiters are comma, semicolon, tab (t), and house (‘ ‘).
This flexibility of pandas permits you to simply write your DataFrame to a CSV file that fits your wants, whether or not it is a typical CSV or a CSV with a particular delimiter.
Writing DataFrame to CSV With out Index
By default, if you write a DataFrame to a CSV file utilizing the to_csv() perform, pandas consists of the DataFrame’s index. Nonetheless, there could also be eventualities the place you don’t need this. In such circumstances, you possibly can set the index parameter to False to exclude the index from the CSV file.
Here is an instance:
import pandas as pd
df = pd.DataFrame({
'A': ['foo', 'bar', 'baz'],
'B': ['alpha', 'beta', 'gamma']
})
print(df)
df.to_csv('no_index.csv', index=False)
The print(df) command will output:
Try our hands-on, sensible information to studying Git, with best-practices, industry-accepted requirements, and included cheat sheet. Cease Googling Git instructions and really be taught it!
A B
0 foo alpha
1 bar beta
2 baz gamma
However the no_index.csv file will appear like this:
A,B
foo,alpha
bar,beta
baz,gamma
As you possibly can see, the CSV file doesn’t embody the DataFrame’s index.
In the event you open the CSV file in a textual content editor, chances are you’ll not see the DataFrame’s index. Nonetheless, should you open the CSV file in a spreadsheet program like Excel, you will notice the index as the primary column.
Dealing with Particular Circumstances
There are just a few particular circumstances chances are you’ll come throughout when writing a DataFrame to a CSV file.
Dealing with NaN Values
By default, pandas will write NaN values to the CSV file. Nonetheless, you possibly can change this habits utilizing the na_rep parameter. This parameter permits you to specify a string that can exchange NaN values.
Here is an instance:
import pandas as pd
import numpy as np
df = pd.DataFrame({
'A': ['foo', np.nan, 'baz'],
'B': ['alpha', 'beta', np.nan]
})
df.to_csv('nan_values.csv', na_rep='NULL')
Within the nan_values.csv file, NaN values are changed with NULL:
,A,B
0,foo,alpha
1,NULL,beta
2,baz,NULL
Writing a Subset of the DataFrame to CSV
Generally, chances are you’ll wish to write solely a subset of the DataFrame to the CSV file. You are able to do this utilizing the columns parameter. This parameter permits you to specify an inventory of column names that you just wish to embody within the CSV file.
Here is an instance:
import pandas as pd
df = pd.DataFrame({
'A': ['foo', 'bar', 'baz'],
'B': ['alpha', 'beta', 'gamma'],
'C': [1, 2, 3]
})
df.to_csv('subset.csv', columns=['A', 'B'])
The subset.csv file will embody solely the ‘A’ and ‘B’ columns:
,A,B
0,foo,alpha
1,bar,beta
2,baz,gamma
Bear in mind, pandas is a strong library and supplies many choices for writing DataFrames to CSV information. Remember to take a look at the official documentation to be taught extra.
Conclusion
On this tutorial, we now have explored the facility of pandas and its potential to write down DataFrame to a CSV file. We have discovered the essential methodology of writing a DataFrame to a CSV file, learn how to specify a delimiter, and learn how to write a DataFrame to a CSV file with out the index. We have additionally checked out dealing with particular circumstances in writing a DataFrame to a CSV file.
