Pandas: High-Performance Evaluation with eval()

Evaluate mathematical expressions efficiently on Pandas DataFrames using the df.eval() method. Perform high-performance column computations, assign new columns inplace, and reference local Python variables using the @ prefix.

Show Table of Contents ↓

Introduction to eval() 🔝

The df.eval() method evaluates Python expression strings using underlying engine optimizations (such as numexpr when installed). It accelerates large numerical computations on DataFrame columns while keeping code clean and readable.

Create numerical sample dataset 🔝

We start with a sales transaction DataFrame containing quantities and unit prices.

import pandas as pd

df = pd.DataFrame({
    "ITEM": ["Apple", "Banana", "Cherry", "Date"],
    "QTY": [10, 25, 15, 30],
    "PRICE": [20.0, 5.0, 50.0, 12.0]
})
print(df)
assert df.shape == (4, 3)

Expected output

     ITEM  QTY  PRICE
0   Apple   10   20.0
1  Banana   25    5.0
2  Cherry   15   50.0
3    Date   30   12.0

Basic column evaluation 🔝

Pass a mathematical expression string to df.eval() to compute operations across columns without explicit bracket indexers.

res = df.eval("QTY * PRICE")
print(res)
assert res.iloc[0] == 200.0

Expected output

0    200.0
1    125.0
2    750.0
3    360.0
dtype: float64

Modifying DataFrames with inplace=True 🔝

Use inplace=True to create and assign a new column directly inside the existing DataFrame.

df.eval("TOTAL = QTY * PRICE", inplace=True)
print(df)
assert "TOTAL" in df.columns and df.loc[0, "TOTAL"] == 200.0

Expected output

     ITEM  QTY  PRICE  TOTAL
0   Apple   10   20.0  200.0
1  Banana   25    5.0  125.0
2  Cherry   15   50.0  750.0
3    Date   30   12.0  360.0

Referencing local variables using @ 🔝

Prefix local Python variables with @ inside your expression string to combine external parameters with DataFrame columns.

tax_rate = 0.1
df.eval("TAX = TOTAL * @tax_rate", inplace=True)
print(df)
assert df.loc[0, "TAX"] == 20.0

Expected output

     ITEM  QTY  PRICE  TOTAL   TAX
0   Apple   10   20.0  200.0  20.0
1  Banana   25    5.0  125.0  12.5
2  Cherry   15   50.0  750.0  75.0
3    Date   30   12.0  360.0  36.0

Exercise: compute net sales value 🔝

Create a DataFrame with PRICE = [100, 200, 300] and DISCOUNT = [10, 20, 30]. Use df.eval() to compute a new column NET = PRICE - DISCOUNT and verify that the first row net equals 90.

Exercise solution 🔝

Solution constructing the DataFrame and applying eval().

ex_df = pd.DataFrame({"PRICE": [100, 200, 300], "DISCOUNT": [10, 20, 30]})
ex_df.eval("NET = PRICE - DISCOUNT", inplace=True)
print(ex_df)
assert ex_df.loc[0, "NET"] == 90

Expected output

   PRICE  DISCOUNT  NET
0    100        10   90
1    200        20  180
2    300        30  270

Practice in Google Colab 🔝

Interactive Colab Notebook

Practice eval() examples interactively directly in your browser:

View on GitHub

Reference: Pandas eval() documentation.

Subhendu Mohapatra

Author & Instructor at plus2net

I write and maintain practical tutorials on Python, PHP, SQL, JavaScript, HTML, jQuery, and web development at plus2net. The tutorials focus on clear explanations, working examples, and code that readers can test and adapt while learning.

About the author Watch on YouTube

Connect:

Subscribe to our YouTube Channel here