箱线图和数据点在一个图中并排答案

【问题标题】：Boxplot and Data points side by side in one plot箱线图和数据点在一个图中并排
【发布时间】：2020-05-06 23:25:23
【问题描述】：

plt.figure(figsize=(8,5))
sns.boxplot(x=df.StoreType, y=df.Sales)

我得到了上面的图，但我想要并排的箱线图和数据点（不重叠，使用 seaborn 或 matplotlib），如下所示：

【问题讨论】：

标签： python matplotlib seaborn

【解决方案1】：

下面的代码借鉴了其他几个 SO 答案：

抵消swarmplot的思路来自：https://stackoverflow.com/a/56655927/42346
改变swarmplot的宽度的代码来自：https://stackoverflow.com/a/44542112/42346

如果您对代码有任何疑问，请告诉我。

import seaborn as sns, numpy as np
import matplotlib, matplotlib.pyplot as plt

tips = sns.load_dataset('tips')

# adjust these as per your data 
boxplot_width = .25 # thinner to make room for having swarmplot beside
swarmplot_offset = -.5 # offset to left of boxplot
xlim_offset = -1 # necessary to show leftmost swarmplot  

fig = plt.figure(figsize=(6,4))
ax = sns.swarmplot(x="day", y="total_bill", data=tips)

path_collections = [child for child in ax.get_children() 
                    if isinstance(child,matplotlib.collections.PathCollection)] 

for path_collection in path_collections: 
    x,y = np.array(path_collection.get_offsets()).T 
    xnew = x + swarmplot_offset
    offsets = list(zip(xnew,y)) 
    path_collection.set_offsets(offsets)

sns.boxplot(x="day", y="total_bill", data=tips, width=boxplot_width, ax=ax) 

def change_width(ax, new_value):
    for patch in ax.patches:
        current_width = patch.get_width()
        diff = current_width - new_value

        # change patch width
        patch.set_width(new_value)

        # re-center patch
        patch.set_x(patch.get_x() + diff * .5)

change_width(ax,.25)

ax.set_xticklabels(ax.get_xticklabels(), ha="right") # align labels to left
ax.set_xlim(xlim_offset,ax.get_xlim()[1]) # to show leftmost swarmplot

plt.show()

示例图片：

【讨论】：