Bokeh多级分类条形图官方示例代码无法运行问题咨询

Hey there! Let’s work through that frustrating issue with getting Bokeh’s multi-level categorical bar chart example up and running. I’ve hit similar roadblocks before, so here’s what to check and a full, working version of the code to get you started:

1. First, Check Your Dependencies & Bokeh Version #

Outdated packages are the #1 culprit for these kinds of problems. Make sure you’ve got the latest stable Bokeh and pandas (most official examples rely on it) installed:

pip install --upgrade bokeh pandas

Older Bokeh versions (pre-2.0) had different syntax for FactorRange and categorical mappings, so upgrading will eliminate most compatibility bugs.

2. Use a Complete, Self-Contained Code Snippet #

Sometimes the official docs skip small but critical bits (like imports or data setup) in their example snippets. Here’s a fully runnable version of the multi-level bar chart that matches the user guide example:

from bokeh.io import show, output_file
from bokeh.models import ColumnDataSource, FactorRange
from bokeh.plotting import figure
from bokeh.transform import factor_cmap
from bokeh.palettes import Viridis256

# Set up multi-category data
fruits = ['Apples', 'Pears', 'Nectarines', 'Plums', 'Grapes', 'Strawberries']
years = ['2015', '2016', '2017']

# Build nested factor tuples for the x-axis
x = [(fruit, year) for fruit in fruits for year in years]
counts = [
    4, 5, 8, 12, 8, 9,
    2, 3, 4, 8, 7, 6,
    1, 2, 4, 4, 5, 4
]

# Create data source
source = ColumnDataSource(data=dict(x=x, counts=counts))

# Initialize plot with FactorRange
p = figure(
    x_range=FactorRange(*x), 
    height=350, 
    title="Fruit Counts by Year",
    toolbar_location=None, 
    tools=""
)

# Add vertical bars with color mapping
p.vbar(
    x='x', 
    top='counts', 
    width=0.9, 
    source=source,
    fill_color=factor_cmap(
        'x', 
        palette=Viridis256, 
        factors=years, 
        start=1, 
        end=2
    )
)

# Clean up the plot appearance
p.y_range.start = 0
p.x_range.range_padding = 0.1
p.xaxis.major_label_orientation = 1  # Rotate labels for readability
p.xgrid.grid_line_color = None

# Output to HTML (or just use show() to display instantly)
output_file("multi_level_bars.html")
show(p)

3. Fix Version-Specific Syntax Issues #

If you’re stuck on an older Bokeh version (2.x instead of 3.x), adjust the factor_cmap line to remove the start and end parameters—those were added in 3.x to handle nested factors more explicitly:

fill_color=factor_cmap('x', palette=Viridis256, factors=years)

If you’re still getting errors, share the exact error message you’re seeing (e.g., import errors, runtime exceptions, or blank plots) and we can narrow it down further!

内容的提问来源于stack exchange，提问作者Matt P.