DataFrame API in Snowpark in Snowflake - Mini Project: Build & Apply

DataFrame API in Snowpark

📖 Scenario: You are working with Snowflake's Snowpark to process and analyze sales data stored in a table. You want to use the DataFrame API to manipulate this data easily without writing complex SQL queries.

🎯 Goal: Build a Snowpark DataFrame pipeline that loads sales data, filters for sales above a certain amount, selects specific columns, and orders the results by sale date.

📋 What You'll Learn

Create a Snowpark DataFrame from the sales_data table

Add a filter to select sales where amount is greater than 1000

Select the columns sale_id, customer_id, and amount

Order the results by sale_date in descending order

💡 Why This Matters

🌍 Real World

Data analysts and engineers use Snowpark DataFrames to process large datasets in Snowflake without writing complex SQL, making data workflows easier and more maintainable.

💼 Career

Understanding Snowpark DataFrame API is valuable for roles involving cloud data engineering, data analysis, and building scalable data pipelines on Snowflake.

1

Create initial DataFrame from sales_data table

Create a Snowpark DataFrame called df by reading the table named sales_data using the session object session.

Snowflake

# Create a DataFrame called df from the sales_data table
# Your code here

Need a hint?

Use session.table("sales_data") to create the DataFrame.

2

Filter sales with amount greater than 1000

Add a filter to the DataFrame df to keep only rows where the amount column is greater than 1000. Assign the result back to df.

Snowflake

df = session.table("sales_data")
# Filter df where amount > 1000
# Your code here

Need a hint?

Use df.filter(df["amount"] > 1000) to filter the DataFrame.

3

Select specific columns

Select the columns sale_id, customer_id, and amount from the DataFrame df. Assign the result back to df.

Snowflake

df = session.table("sales_data")
df = df.filter(df["amount"] > 1000)
# Select sale_id, customer_id, and amount columns
# Your code here

Need a hint?

Use df.select("sale_id", "customer_id", "amount") to select columns.

4

Order results by sale_date descending

Order the DataFrame df by the sale_date column in descending order. Assign the result back to df.

Snowflake

df = session.table("sales_data")
df = df.filter(df["amount"] > 1000)
df = df.select("sale_id", "customer_id", "amount")
# Order df by sale_date descending
# Your code here

Need a hint?

Use df.order_by(df["sale_date"].desc()) to order descending.