Devendra174's picture
Upload folder using huggingface_hub
1e92f2d verified
raw
history blame
2.03 kB
flavors:
- flavor: svg
path: /swarmplot/
- flavor: canvas
path: /swarmplot/canvas/
SwarmPlot:
package: '@nivo/swarmplot'
tags: []
stories:
- label: Custom circle component
link: swarmplot--custom-circle-component
- label: Adding extra layers
link: swarmplot--extra-layers
- label: Using annotations
link: swarmplot--using-annotations
description: |
A swarm plot component which can also be used to make a beeswarm plot
when using a single group.
This chart can display 2 data dimensions, a categorical one: **groups**,
and a quantitative one: **values**.
You can optionally add a third quantitative dimension if you enable
**dynamic node size**, please have a look at the `size` property
for further information.
You can also enable a voronoi mesh to capture user's
interactions, using the `useMesh` property.
This example uses 3 dimensions, the grouping is done by the `group`
property while the value is determined by the `price` property,
node size is dynamic and depends on the `volume` property.
Under the hood, this chart uses [d3-force](https://github.com/d3/d3-force)
with a simulation involving collisions and x/y forces, the quality of
the simulation can be adjusted, the strength of the *value* force too.
SwarmPlotCanvas:
package: '@nivo/swarmplot'
tags:
- canvas
stories:
- label: Using annotations
link: swarmplotcanvas--using-annotations
description: |
A variation around the [SwarmPlot](self:/swarmplot) component.
Well suited for large data sets as it does not impact DOM tree depth,
however you'll lose the isomorphic ability and transitions.
You can optionally enable a voronoi mesh to capture user's
interactions, using the `useMesh` property.
Even if the canvas implementation is faster, please note that
if you have a lot of nodes **calculating the underlying simulation
will involve a lot of computing and will affect performance**.