问题
I have a GlyphRenderer
whose data_source.data is
{'index': [0, 1, 2, 3, 4, 5, 6, 7],
'color': ['#3288bd', '#66c2a5', '#abdda4', '#e6f598', '#fee08b', '#fdae61', '#f46d43', '#d53e4f']}
The renderer's glyph is
Oval(height=0.1, width=0.2, fill_color="color")
When rendering, I see
E-1001 (BAD_COLUMN_NAME): Glyph refers to nonexistent column name: color [renderer: GlyphRenderer(id='1d1031f5-6ee3-4744-a0f7-22309798e313', ...)]
I'm clearly missing something, but this is pretty much lifted from published examples. I verified in a debugger that data_source.column_names
is just ['index']
; what I don't understand is why the 'color'
column doesn't appear in the data source's column_names
, or why Bokeh produces this warning (the graph appears to be correctly rendered).
The complete source is available at https://pastebin.com/HXAEEujP
回答1:
It's generally better to provide all relevant arguments when constructing an object rather than mutating the object after it's already been created. It's especially true for Bokeh - in many cases it does some additional work based on the arguments passed to __init__
.
Take a look at this version of your code:
import math
from bokeh.io import show
from bokeh.models import GraphRenderer, StaticLayoutProvider, Oval, GlyphRenderer, ColumnDataSource, MultiLine
from bokeh.palettes import Spectral8
from bokeh.plotting import figure
N = 8
node_indices = list(range(N))
plot = figure(title="Graph Layout Demonstration", x_range=(-1.1, 1.1), y_range=(-1.1, 1.1),
plot_width=250, plot_height=250,
tools="", toolbar_location=None)
node_ds = ColumnDataSource(data=dict(index=node_indices,
color=Spectral8),
name="Node Renderer")
edge_ds = ColumnDataSource(data=dict(start=[0] * N,
end=node_indices),
name="Edge Renderer")
### start of layout code
circ = [i * 2 * math.pi / 8 for i in node_indices]
x = [math.cos(i) for i in circ]
y = [math.sin(i) for i in circ]
graph_layout = dict(zip(node_indices, zip(x, y)))
graph = GraphRenderer(node_renderer=GlyphRenderer(glyph=Oval(height=0.1, width=0.2, fill_color="color"),
data_source=node_ds),
edge_renderer=GlyphRenderer(glyph=MultiLine(),
data_source=edge_ds),
layout_provider=StaticLayoutProvider(graph_layout=graph_layout))
plot.renderers.append(graph)
show(plot)
来源:https://stackoverflow.com/questions/49358586/jupyter-bokeh-non-existent-column-name-in-glyph-renderer