Notebook Quickstart¶

Strata Notebook is an interactive notebook with content-addressed caching, automatic dependency tracking, and cascade execution.

1. Start the Server¶

DockerFrom source

docker compose up -d --build

uv run strata-server

Click New Notebook on the landing page. Choose a name and parent directory.

Each notebook gets its own Python environment (managed by uv), so packages installed in one notebook don't affect others.

Add a cell and type:

x = 1
x + 1

Press ++shift+enter++ to run. The result appears below the cell.

Add a second cell:

y = x * 10
print(f"y = {y}")

Strata automatically detects that this cell references x from the first cell. The DAG in the sidebar shows the dependency arrow.

Cascade execution

If you change the first cell and re-run the second, Strata detects the staleness and offers to cascade-execute both cells in the correct order.

The Markdown helper is injected into every cell's namespace:

Markdown("# Hello\n\n- item one\n- item two")

import matplotlib.pyplot as plt

plt.plot([1, 2, 3], [1, 4, 9])
plt.show()

Install matplotlib first

Open the Environment panel in the sidebar and add matplotlib before running plot cells.

A cell can produce multiple visible outputs:

display(Markdown("## Summary"))
42

import pandas as pd

df = pd.DataFrame({
    "name": ["Alice", "Bob", "Carol"],
    "score": [95, 87, 92],
})
df

DataFrames render as scrollable tables with column headers and row counts.

Open the Environment panel in the sidebar to:

When you re-run a cell with the same inputs and source, Strata returns the cached result instantly. The cell shows a ⚡ cached badge with timing.

Change the source or any upstream cell, and the cache is automatically invalidated.

Example notebooks must live under the configured notebook storage root (STRATA_NOTEBOOK_STORAGE_DIR, default /tmp/strata-notebooks).

Copy one of the bundled examples there first:

cp -R examples/iris_classification /tmp/strata-notebooks/

Then use Open Existing and open:

/tmp/strata-notebooks/iris_classification

Other bundled examples:

examples/pandas_basics
examples/titanic_ml