Polars Sum Multiple Columns

In Polars, you can sum multiple columns either row-wise or column-wise using the sum() function along with the select() or with_columns() method, depending on your requirements. In this article, I will explain summing multiple columns in Polars.

Key Points –

You can sum multiple columns row-wise using the + operator (df.with_columns((pl.col("A") + pl.col("B")).alias("Total"))).
The sum_horizontal() function allows summing all numeric columns dynamically without specifying each column manually.
To sum specific columns downward, use df.select(pl.col(["A", "B"]).sum()).
df.select(pl.all().sum()) computes the sum of all columns, including non-numeric ones if not excluded.
Always use alias("New Column Name") when adding sum results to a DataFrame to keep things organized.
pl.sum_horizontal() is more efficient for row-wise operations than using pl.concat_list().

Usage of Polars Sum Multiple Columns

The sum operation in Polars is used to compute the total of multiple columns either row-wise (horizontally) or column-wise (vertically).

To run some examples of the Polars sum multiple columns, let’s create a Polars DataFrame.


import polars as pl

studentdetails = {
       "Studentname":["Ram", "Sam", "Scott", "Ann", "John"],
       "Mathematics" :[80,90,85,72,95],
       "Science" :[85,95,80,90,92],
       "English" :[90,85,80,75,95]
              }
df = pl.DataFrame(studentdetails)
print("Original DataFrame:\n", df)

Yields below output.

You can sum multiple columns row-wise (for each row) using the + operator inside with_columns(). This method is explicit and useful when you want to sum specific columns.


# Sum the subject marks row-wise using the + operator
df2 = df.with_columns((pl.col("Mathematics") + pl.col("Science") + pl.col("English")).alias("Total Marks"))
print(df2)

Here,

The + operator provides an easy-to-read method for summing columns row-wise.
We manually add each column (Mathematics, Science, and English) using +.
The result is stored in a new column "Total Marks".
The sum is calculated row-wise for each student.

Sum Columns Row-wise Using pl.concat_list() Function

The pl.concat_list() function in Polars allows us to concatenate multiple column values into a list and then apply functions like sum() to compute the row-wise sum dynamically.


# Sum multiple columns row-wise using pl.concat_list()
df2 = df.with_columns(pl.concat_list(["Mathematics", "Science", "English"]).list.sum().alias("Total Marks"))
print(df2)

Here,

pl.concat_list(["Mathematics", "Science", "English"]) Combines selected columns into a list.
.list.sum() Computes the sum for each row.
.alias("Total Score") Renames the result as "Total Marks".

Yields same output as above.

Sum All Numeric Columns Row-wise

When summing all numeric columns row-wise in Polars, we can use the sum_horizontal() function, which dynamically sums all numeric columns without needing to specify them manually.


# Sum all numeric columns row-wise
df2 = df.with_columns(pl.sum_horizontal(pl.all().exclude("Studentname")).alias("Total Marks"))
print(df2)

# Sum all numeric columns row-wise
df2 = df.with_columns(pl.sum_horizontal(pl.exclude("Studentname")).alias("Total Marks"))
print(df2)

Here,

pl.exclude("Studentname") Excludes non-numeric columns.
pl.sum_horizontal() Sums all remaining numeric columns row-wise.
.alias("Total Score") Renames the result as "Total Marks".

Yields same output as above.

Sum Specific Columns Column-Wise

To sum specific columns column-wise (i.e., computing the sum for each selected column) in Polars, you can use the select() method along with pl.col().sum() function.


# Sum specific columns column-wise
df2 = df.select(pl.col(["Mathematics", "Science"]).sum())
print(df2)

# Output:
# shape: (1, 2)
┌─────────────┬─────────┐
│ Mathematics ┆ Science │
│ ---         ┆ ---     │
│ i64         ┆ i64     │
╞═════════════╪═════════╡
│ 422         ┆ 442     │
└─────────────┴─────────┘

Here,

pl.col(["Mathematics", "Science"]).sum() Selects only Mathematics and Science, then computes their column-wise sum.
select() Extracts only the computed sum values.
The output is a new single-row DataFrame with summed values.

Sum Multiple Columns Column-Wise

To sum multiple columns column-wise (i.e., computing the sum for each column individually) in Polars, you can use select(pl.all().sum()) or sum() on selected columns.


# Sum multiple columns downward (column-wise)
df2 = df.select(pl.col(["Mathematics", "Science", "English"]).sum())
print(df2)

# Sum multiple columns column-wise
df2 = df.select(pl.all().exclude("Studentname").sum())
print(df2)

# Output:
# shape: (1, 3)
┌─────────────┬─────────┬─────────┐
│ Mathematics ┆ Science ┆ English │
│ ---         ┆ ---     ┆ ---     │
│ i64         ┆ i64     ┆ i64     │
╞═════════════╪═════════╪═════════╡
│ 422         ┆ 442     ┆ 425     │
└─────────────┴─────────┴─────────┘

Here,

pl.col(["Mathematics", "Science", "English"]).sum() selects the specified columns and calculates the column-wise sum.
select() returns a single-row DataFrame where each column holds its sum.

Conclusion

In conclusion, polars provides efficient and flexible methods to sum multiple columns, both row-wise and column-wise. Using functions like the + operator, sum_horizontal(), pl.col().sum(), pl.concat_list(), and pl.exclude(), you can dynamically compute sums without the need to manually specify column names.

Happy Learning!!

References

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

Polars Sum Multiple Columns

Usage of Polars Sum Multiple Columns

Sum Columns Row-wise Using pl.concat_list() Function

Sum All Numeric Columns Row-wise

Sum Specific Columns Column-Wise

Sum Multiple Columns Column-Wise

Conclusion

Related Articles

References