Polars DataFrame rows() Method with Examples

In Polars, the rows() method converts a DataFrame into native Python data structures. By default, it returns a list of tuples, with each tuple representing a row in the DataFrame. If named=True is passed, it returns a list of dictionaries instead, using column names as keys.

Polars DataFrame rows() Introduction

Let’s know the syntax of the DataFrame rows() function


# Syntax of DataFrame rows()
DataFrame.rows(
*,
named: bool = False,
) → list[tuple[Any, ...]] | list[dict[str, Any]]

Parameters of the DataFrame rows()

Following are the parameters of the DataFrame rows() function

named: bool = False (optional keyword-only argument)
- If False (default), returns a list of tuples (each tuple is a row).
- If True, returns a list of dictionaries (each dict maps column names to row values).

Return Value

This function returns a list of tuples, where each tuple represents a row in the DataFrame.

Usage of Polars DataFrame rows() Method

It returns all rows of the DataFrame as either a list of tuples or a list of dictionaries, depending on the named parameter.

Now, let’s create a Polars DataFrame.


import polars as pl

technologies= {
    'Courses':["Spark","PySpark","Hadoop","Python"],
    'Fees' :[22000,25000,23000,24000],
    'Duration':['30days','50days','60days','40days'],
    'Discount':[1000,2000,1200,2500]
          }
df = pl.DataFrame(technologies)
print("Original DataFrame:\n", df)

Yields below output.

To retrieve the rows of your Polars DataFrame as a list of tuples (the default output of the rows() method), simply call rows() on your DataFrame.


# Get rows as list of tuples
rows_list = df.rows()
print(rows_list)

In the above example, each tuple in the list represents a row of the DataFrame, with values in the order of columns: Courses, Fees, Duration, Discount.

Get Rows as a List of Dictionaries (Named=True)

To retrieve the rows of a Polars DataFrame as a list of dictionaries, use the rows(named=True) method. Each row will be represented as a dictionary, with column names as keys and the corresponding values as dictionary values.


# Get rows as a list of dictionaries
rows_as_dicts = df.rows(named=True)
print(rows_as_dicts)

# Output:
# [
# {'Courses': 'Spark', 'Fees': 22000, 'Duration': '30days', 'Discount': 1000},
# {'Courses': 'PySpark', 'Fees': 25000, 'Duration': '50days', 'Discount': 2000},
# {'Courses': 'Hadoop', 'Fees': 23000, 'Duration': '60days', 'Discount': 1200},
# {'Courses': 'Python', 'Fees': 24000, 'Duration': '40days', 'Discount': 2500}
# ]

In the above example, rows() returns a list of tuples, while rows(named=True) returns a list of dictionaries where column names serve as keys, handy when you need named access to each value.

Iterate Over Rows as Tuples

To iterate over rows as tuples in a Polars DataFrame, use a for loop in combination with the rows() method.


# Iterate over rows as tuples
for row in df.rows():
    print

# Output:
# ('Spark', 22000, '30days', 1000)
# ('PySpark', 25000, '50days', 2000)
# ('Hadoop', 23000, '60days', 1200)
# ('Python', 24000, '40days', 2500)

In the above example, Each row is a Python tuple, with values in the same order as the columns in the DataFrame.

Iterate Over Rows as Dictionaries

To iterate over rows as dictionaries in a Polars DataFrame, use the rows(named=True) method within a loop. This returns each row as a dictionary where the keys are the column names.


# Iterate over rows as dictionaries
for row_dict in df.rows(named=True):
    print(row_dict)

# Output:
# {'Courses': 'Spark', 'Fees': 22000, 'Duration': '30days', 'Discount': 1000}
# {'Courses': 'PySpark', 'Fees': 25000, 'Duration': '50days', 'Discount': 2000}
# {'Courses': 'Hadoop', 'Fees': 23000, 'Duration': '60days', 'Discount': 1200}
# {'Courses': 'Python', 'Fees': 24000, 'Duration': '40days', 'Discount': 2500}

In the above example, each row_dict is a dictionary where keys are column names and values are the row values. Perfect for more readable or key-based access inside your loop.

Access First Row Tuple

To access the first row as a tuple from a Polars DataFrame, you can use the rows() method and then index the first element [0].


# Get the first row as a tuple
df2 = df.rows()[0]
print(df2)

# Output:
# ('Spark', 22000, '30days', 1000)

In the above example, this returns the first row as a tuple with all column values in order.

Access First Row as Dict

To access the first row as a dictionary in a Polars DataFrame, call rows(named=True) and retrieve the first element [0] from the resulting list.


# Get the first row as a dictionary
df2 = df.rows(named=True)[0]
print(df2)

# Output:
# {'Courses': 'Spark', 'Fees': 22000, 'Duration': '30days', 'Discount': 1000}

In the above example, this gives you the first row with column names as keys and the corresponding values.

Conclusion

In summary, Polars provides flexible methods to access DataFrame rows, whether as tuples using rows() or as dictionaries with rows(named=True). These options make it easy to iterate over and manipulate data in the format that best suits your needs.

Happy Learning!!

Reference

https://docs.pola.rs/api/python/stable/reference/dataframe/api/polars.DataFrame.sample.html