Quickstart

Getting Started

Install ipydeck into your environment. The package ships as a standard Python wheel, so any installer that understands pyproject.toml metadata will work.

pip install ipydeck

If you prefer using uv you can instead run:

uv add ipydeck

Create Your First Deck

Launch a Jupyter notebook and construct a Deck widget. The example below renders a scatterplot centered on San Francisco.

from ipydeck import Deck, Layer, ViewState

scatter = Layer(
    type="ScatterplotLayer",
    data=[{"position": [-122.4, 37.8], "radius": 1200}],
    get_position="@@=position",
    get_radius="@@=radius",
    get_fill_color=[0, 0, 255],
    pickable=True,
)

deck = Deck(
    layers=[scatter],
    initial_view_state=ViewState(latitude=37.8, longitude=-122.4, zoom=11),
    tooltip={"text": "deck.gl"},
)

deck

Interacting with the Widget

Updating layer properties updates the rendered map after calling Deck.update(). This is useful for stateful UIs where the same layer instance is mutated in-place. The example shows the refreshed serialized payload.

deck2 = Deck(
    layers=[scatter],
    initial_view_state=ViewState(latitude=37.8, longitude=-122.4, zoom=11),
    tooltip={"text": "deck.gl"},
)
deck2.layers[0].data[0]["radius"] = 200
deck2.update()
deck2

The Layer, ViewState, and Deck API references contain additional options that match the underlying deck.gl library.