Is it possible to change the edge color of a node in a Sankey diagram?

tommycarstensen · November 23, 2022, 11:19am

In the Sankey diagram examples on this page the edge color of the nodes is set to black by default:
Sankey — HoloViews v1.15.2

Is it possible to specify a custom color similar to what has been done for the edge color of the flow?

Thanks!

ivandorte · November 23, 2022, 7:48pm

Hi @tommycarstensen,
Maybe your are looking for the node_line_color option:

import holoviews as hv
from holoviews import opts, dim
hv.extension('bokeh')

sankey = hv.Sankey([
    ['A', 'X', 5],
    ['A', 'Y', 7],
    ['A', 'Z', 6],
    ['B', 'X', 2],
    ['B', 'Y', 9],
    ['B', 'Z', 4]]
)

sankey.opts(node_line_color="red",
            width=600,
            height=400)

tommycarstensen · November 24, 2022, 8:59am

That’s exactly what I was looking for. Thank you very much!

LinuxIsCool · November 15, 2023, 12:52am

Where did you find that? I don’t see node_line_color documented anywhere in the hvplot documentation.

ivandorte · November 15, 2023, 11:42am

From this example about Network Graphs:

github.com

holoviz/holoviews/blob/a3b419a22e77a7835cc36dbda0c5ce318dcb4b8f/examples/user_guide/Network_Graphs.ipynb#L363


      
           "source": [
            "kwargs = dict(width=800, height=800, xaxis=None, yaxis=None)\n",
            "opts.defaults(opts.Nodes(**kwargs), opts.Graph(**kwargs))\n",
            "\n",
            "colors = ['#000000']+hv.Cycle('Category20').values\n",
            "edges_df = pd.read_csv('../assets/fb_edges.csv')\n",
            "fb_nodes = hv.Nodes(pd.read_csv('../assets/fb_nodes.csv')).sort()\n",
            "fb_graph = hv.Graph((edges_df, fb_nodes), label='Facebook Circles')\n",
            "\n",
            "fb_graph.opts(cmap=colors, node_size=10, edge_line_width=1,\n",
            "              node_line_color='gray', node_color='circle')"
           ]
          },
          {
           "cell_type": "markdown",
           "metadata": {},
           "source": [
            "## Bundling graphs"
           ]
          },
          {

Sankey plots depends on hv.Graph:

github.com

holoviz/holoviews/blob/a3b419a22e77a7835cc36dbda0c5ce318dcb4b8f/holoviews/plotting/bokeh/init.py#L293


      
          options.HSpan = Options('style', color=Cycle(), alpha=0.5)
          options.Arrow = Options('style', arrow_size=10)
          options.Labels = Options('style', text_align='center', text_baseline='middle')
          options.HLines = Options('style', color=Cycle(), line_width=3, alpha=1, muted_alpha=0.2)
          options.VLines = Options('style', color=Cycle(), line_width=3, alpha=1, muted_alpha=0.2)
          options.VSpans = Options('style', color=Cycle(), alpha=0.5, muted_alpha=0.2)
          options.HSpans = Options('style', color=Cycle(), alpha=0.5, muted_alpha=0.2)
          options.Labels = Options('style', text_color=Cycle(), text_align='center', text_baseline='middle')
          
          # Graphs
          options.Graph = Options(
              'style', node_size=15, node_color=Cycle(), node_line_color='black',
              node_nonselection_fill_color=Cycle(), node_hover_line_color='black',
              node_hover_fill_color='limegreen', node_nonselection_alpha=0.2,
              edge_nonselection_alpha=0.2, node_nonselection_line_color='black',
              edge_color='black', edge_line_width=2, edge_nonselection_line_color='black',
              edge_hover_line_color='limegreen'
          )
          options.TriMesh = Options(
              'style', node_size=5, node_line_color='black', node_color='white',
              edge_line_color='black', node_hover_fill_color='limegreen',