Datashading with dynamical update

[mb-915]

Hello, I'm trying to use datashader with. I use hvplot for this with datashade. Since datashading is not provided by pyodide, I used downsample instead for providing a working sample.

What I don't understand is that in the WASM playground it works: the sampling is updated when you zoom in/out. But on the same local notebook (marimo 0.11.2), it does not : points are not updated when you zoom in/out

Does anyone know how to solve or diagnostic this issue I have with datashading and hvplot ?

p.s.: if there is a way to do rasterization/datashading with altair/vega I would love to use that. But I could not understand if and how VegaFusion is supposed to work for that usecase.

Samples

https://marimo.app/l/legm4e

The native notebook with both datashading and downsampling

import marimo

__generated_with = "0.11.2"
app = marimo.App()


@app.cell
def _():
    import marimo as mo
    import hvplot.pandas
    import pandas as pd
    import numpy as np
    return hvplot, mo, np, pd


@app.cell
def _(np):
    rng = np.random.default_rng(12345)
    x = rng.normal(size=1_000_000)
    y = rng.normal(size=1_000_000)
    x, y
    return rng, x, y


@app.cell
def _(pd, x, y):
    my_points = pd.DataFrame({"x": x, "y": y})

    my_points.hvplot.points(downsample=True)
    return (my_points,)


@app.cell
def _(my_points):
    my_points.hvplot.points(datashade=True, dynspread=True)
    return


if __name__ == "__main__":
    app.run()

Those two figures are not updated when you zoom in/out.

[mb-915]

related issue : #3780