基于Bokeh的单数据列箱线图生成问题：现有分组箱线图代码无法适配的解决建议

The issue with your current code is that it's hardcoded to rely on grouping by a categorical variable. To create a box plot for a single column (like math score) showing its overall distribution, we need to adjust the function to handle a "single group" scenario where we treat the entire dataset as one category.

Here's the modified version of your function that supports both grouped and single-column box plots:

from bokeh.models import Range1d
from bokeh.plotting import figure, show

def box_plot(df, vals, label=None, ylabel=None, xlabel=None, title=None):
    # Handle single column (no grouping) case
    if label is None:
        # Create a dummy group with the entire dataset
        df_gb = df.groupby(lambda x: "Overall")
        cats = ["Overall"]
    else:
        # Original grouping logic
        df_gb = df.groupby(label)
        cats = list(df_gb.groups.keys())
        cats = [str(i) for i in cats]
    
    # Compute quartiles for each group
    q1 = df_gb[vals].quantile(q=0.25)
    q2 = df_gb[vals].quantile(q=0.5)
    q3 = df_gb[vals].quantile(q=0.75)
    
    # Compute interquartile region and outlier bounds
    iqr = q3 - q1
    upper_cutoff = q3 + 1.5*iqr
    lower_cutoff = q1 - 1.5*iqr
    
    # Find outliers for each group
    def outliers(group):
        cat = group.name
        outlier_inds = (group[vals] > upper_cutoff[cat]) | (group[vals] < lower_cutoff[cat])
        return group[vals][outlier_inds]
    
    out = df_gb.apply(outliers).dropna()
    
    # Prepare outlier points for plotting
    outx = []
    outy = []
    for cat in cats:
        if cat in out and not out[cat].empty:
            for value in out[cat]:
                outx.append(cat)
                outy.append(value)
    
    # Adjust whiskers to exclude outliers
    qmin = df_gb[vals].min()
    qmax = df_gb[vals].max()
    upper = [min([x,y]) for (x,y) in zip(qmax, upper_cutoff)]
    lower = [max([x,y]) for (x,y) in zip(qmin, lower_cutoff)]
    
    # Build figure
    p = figure(sizing_mode='stretch_width', x_range=cats, height=300, toolbar_location=None)
    p.xgrid.grid_line_color = None
    p.ygrid.grid_line_width = 2
    p.yaxis.axis_label = ylabel if ylabel else vals
    p.xaxis.axis_label = xlabel if xlabel else (label if label else "")
    p.title = title if title else f"Box Plot of {vals}"
    p.y_range.start = 0
    p.title.align = 'center'
    
    # Stems
    p.segment(cats, upper, cats, q3, line_width=2, line_color="black")
    p.segment(cats, lower, cats, q1, line_width=2, line_color="black")
    
    # Boxes - use a single color for single group, or original palette for multiple groups
    fill_colors = ['#a50f15'] if label is None else ['#a50f15', '#de2d26', '#fb6a4a', '#fcae91', '#fee5d9']
    p.rect(cats, (q3 + q1)/2, 0.5, q3 - q1, fill_color=fill_colors, alpha=0.7, line_width=2, line_color="black")
    
    # Median line
    p.rect(cats, q2, 0.5, 0.01, line_color="black", line_width=2)
    
    # Whisker caps
    p.rect(cats, lower, 0.2, 0.01, line_color="black")
    p.rect(cats, upper, 0.2, 0.01, line_color="black")
    
    # Outliers
    p.circle(outx, outy, size=6, color="black")
    
    return p

Key Changes Explained: #

• Added a conditional check for label=None: When you don't pass a categorical label, the function creates a dummy group called "Overall" containing the entire dataset.
• Adjusted the fill color palette to use a single color for the single-group case (you can change this to any color you prefer).
• Improved default labels/title for better readability when no custom text is provided.

How to Use It for Single Column: #

To generate a box plot for math score, simply call the function without passing a label parameter:

# Assuming df is your loaded StudentsPerformance dataset
p = box_plot(df, 'math score', ylabel='Math Score', title='Distribution of Math Scores')
show(p)

Why Your Previous Attempt Failed: #

Setting cats = df['math score'] was incorrect because cats is meant to be the list of category labels (like group names), not the raw data values. For a single column, you only need one category label (like "Overall") to anchor the box plot on the x-axis.

内容的提问来源于stack exchange，提问作者curiouscoder