`scatter`

Draw a cloud of points, optionally colored by a data column. The most common way to show raw samples.

Matplotlib: Axes.scatter. Coordinates are forwarded as scatter(x, y, ...) and every style key is passed through as a keyword argument.

Axes: rectangular (ax) and ternary (axtri).

Coordinates

Key	Required	Meaning
`x`	yes	Horizontal position
`y`	yes	Vertical position
`c`	no	Per-point value mapped through `cmap` to a color

On a ternary axis use left / right / bottom instead of x / y.

Style

These are the Matplotlib scatter keywords you will use most; any other scatter keyword also works.

Key	Type	Purpose
`s`	float or array	Marker size in points²
`marker`	string	Marker style, e.g. `"."`, `"o"`, `"x"`, `"+"`
`c` / `color`	color	Single color (use the `c` coordinate for data-mapped color)
`alpha`	0–1	Opacity
`edgecolor`	color	Marker edge color (`"none"` for no edge)
`linewidths`	float	Marker edge width
`cmap`	string	Colormap (when the `c` coordinate is set)
`vmin` / `vmax`	float	Color scale limits
`zorder`	number	Draw order (higher is on top)

Example

- name: samples
  data:
    - source: df
      transform: [ {sortby: LogL} ]   # draw best-fit points last (on top)
  axes: ax
  method: scatter
  coordinates:
    x: {expr: mass}
    y: {expr: "np.log10(xsec)"}
    c: {expr: LogL}
  style:
    marker: "."
    s: 3
    cmap: jarvis_rainbow2_r
    vmin: -50
    vmax: 0
  colorbar: axc                       # attach the color mapping to the colorbar axis

Notes

A constant color comes from style.color; a data-driven color comes from the c coordinate plus a cmap. To show a colorbar, set the layer's colorbar to a colorbar axis such as axc and use a *cmap style card.
Sort by likelihood (or weight) before drawing so the most important points are on top.