Notebooks

Athena Notebooks bring the power of Jupyter notebooks directly into the Athena platform, enhancing your data analysis and coding experience. These notebooks allow you to conduct in-depth, AI-powered analyses that contain interactive code, equations, visualizations, and models.

​

Key Features

​

Chat with Athena in the sidebar

You can collaborate with Athena while you work. She can “see” your notebook and your currently selected cell. So if you are stuck on something, just pop open the sidebar and ask for help or to write the code for you!

​

Call Athena directly in a notebook cell

In Jupyter notebooks the double percentage signs ”%%” are cell magic commands. You can use Athena directly by using the %%ai athena or %%ai athena -f code command followed by your prompt your request.

%%ai athenaGenerate steps to predict trends in lap times from given F1 race dataset.

%%ai athenaGenerate steps to predict trends in lap times from given F1 race dataset.

%%ai athena -f codeWrite python code to perform Exploratory Data Analysis (EDA):
Perform EDA to understand the distribution of lap times and identify any patterns or anomalies.
Visualize the data using plots such as histograms, scatter plots, and box plots to gain insights into the relationships between different variables.
Analyze correlations between lap times and other features to identify potential predictors.

​

Directly load uploaded Excel/CSV datasets within Notebooks

Effortlessly manage your datasets by uploading Excel or CSV files as Athena assets and loading them into Athena Notebooks with a single command via the chat sidebar.

​

Use the Athena SDK

We provide an SDK that exposes some of Athena’s capabilities. The SDK allows you to generate conversation messages, perform web searches, use tools and workflows, and more. You can use that SDK directly within Athena Notebooks, eliminating the need to enter or setup your API Key.

Athena Notebooks enable you to seamlessly execute saved SQL queries within your notebook environment, allowing for efficient data analysis and workflow integration.

​

Step-by-step Instructions

Use query to answer: Which drivers have scored the most points so far in the 2024 season?

Retrieve the asset_id for @your-query-name and @your-database-name.

Queries in notebooks can be executed either by referencing a saved SQL snippet or by directly running SQL statements against your connected databases.

Create a new Notebook.

from athena.client import Athena
# Setup the client
client = Athena ()
# Execute SQL snippet and load results into a DataFrame
df = client.query.execute_snippet(snippet_asset_id="query-asset-id")
#print the df
print(df)

Create a new Notebook.

from athena.client import Athena
# Setup the client
client = Athena ()
# Execute SQL snippet and load results into a DataFrame
df = client.query.execute_snippet(snippet_asset_id="query-asset-id")
#print the df
print(df)

Create a new Notebook.

# Format the SQL command for better readability
client.query.execute(sql_command=```your-sql-query```, database_asset_ids=['database-asset-id'])