如何使 mpld3 与 seaborn 一起工作（交互式工具提示）

Question

我似乎无法获得由 mpld3 提供支持的交互式工具提示来处理来自 seaborn 的奇妙的 lmplot 类散点图。

我很想知道如何让它工作的任何指示！谢谢！

示例代码：

# I'm running this in an ipython notebook.
%matplotlib inline
import matplotlib.pyplot as plt, mpld3
mpld3.enable_notebook()
import seaborn as sns


N=10
data = pd.DataFrame({"x": np.random.randn(N),
                     "y": np.random.randn(N), 
                     "size": np.random.randint(20,200, size=N),
                     "label": np.arange(N)
                     })


scatter_sns = sns.lmplot("x", "y", 
           scatter_kws={"s": data["size"]},
           robust=False, # slow if true
           data=data, size=8)
fig = plt.gcf()

tooltip = mpld3.plugins.PointLabelTooltip(fig, labels=list(data.label))
mpld3.plugins.connect(fig, tooltip)

mpld3.display(fig)

我得到了 seaborn 情节以及以下错误：

Javascript error adding output!
TypeError: obj.elements is not a function
See your browser Javascript console for more details.

控制台显示：

TypeError: obj.elements is not a function
    at mpld3_TooltipPlugin.draw (https://mpld3.github.io/js/mpld3.v0.2.js:1161:9)
    at mpld3_Figure.draw (https://mpld3.github.io/js/mpld3.v0.2.js:1400:23)
    at Object.mpld3.draw_figure (https://mpld3.github.io/js/mpld3.v0.2.js:18:9)
    at eval (eval at <anonymous> (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231), <anonymous>:14:14)
    at eval (eval at <anonymous> (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231), <anonymous>:15:5)
    at eval (native)
    at Function.x.extend.globalEval (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231)
    at x.fn.extend.domManip (https://mbcomp1:9999/static/components/jquery/jquery.min.js:5:21253)
    at x.fn.extend.append (https://mbcomp1:9999/static/components/jquery/jquery.min.js:5:18822)
    at OutputArea._safe_append (https://mbcomp1:9999/static/notebook/js/outputarea.js:336:26)
outputarea.js:319 Javascript error adding output! TypeError: obj.elements is not a function
    at mpld3_TooltipPlugin.draw (https://mpld3.github.io/js/mpld3.v0.2.js:1161:9)
    at mpld3_Figure.draw (https://mpld3.github.io/js/mpld3.v0.2.js:1400:23)
    at Object.mpld3.draw_figure (https://mpld3.github.io/js/mpld3.v0.2.js:18:9)
    at eval (eval at <anonymous> (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231), <anonymous>:14:14)
    at eval (eval at <anonymous> (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231), <anonymous>:15:5)
    at eval (native)
    at Function.x.extend.globalEval (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231)
    at x.fn.extend.domManip (https://mbcomp1:9999/static/components/jquery/jquery.min.js:5:21253)
    at x.fn.extend.append (https://mbcomp1:9999/static/components/jquery/jquery.min.js:5:18822)
    at OutputArea._safe_append (https://mbcomp1:9999/static/notebook/js/outputarea.js:336:26)
outputarea.js:338 TypeError: obj.elements is not a function
    at mpld3_TooltipPlugin.draw (https://mpld3.github.io/js/mpld3.v0.2.js:1161:9)
    at mpld3_Figure.draw (https://mpld3.github.io/js/mpld3.v0.2.js:1400:23)
    at Object.mpld3.draw_figure (https://mpld3.github.io/js/mpld3.v0.2.js:18:9)
    at eval (eval at <anonymous> (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231), <anonymous>:14:14)
    at eval (eval at <anonymous> (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231), <anonymous>:15:5)
    at eval (native)
    at Function.x.extend.globalEval (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231)
    at x.fn.extend.domManip (https://mbcomp1:9999/static/components/jquery/jquery.min.js:5:21253)
    at x.fn.extend.append (https://mbcomp1:9999/static/components/jquery/jquery.min.js:5:18822)
    at OutputArea._safe_append (https://mbcomp1:9999/static/notebook/js/outputarea.js:336:26)
outputarea.js:319 Javascript error adding output! TypeError: obj.elements is not a function
    at mpld3_TooltipPlugin.draw (https://mpld3.github.io/js/mpld3.v0.2.js:1161:9)
    at mpld3_Figure.draw (https://mpld3.github.io/js/mpld3.v0.2.js:1400:23)
    at Object.mpld3.draw_figure (https://mpld3.github.io/js/mpld3.v0.2.js:18:9)
    at eval (eval at <anonymous> (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231), <anonymous>:14:14)
    at eval (eval at <anonymous> (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231), <anonymous>:15:5)
    at eval (native)
    at Function.x.extend.globalEval (https://mbcomp1:9999/static/components/jquery/jquery.min.js:4:4231)
    at x.fn.extend.domManip (https://mbcomp1:9999/static/components/jquery/jquery.min.js:5:21253)
    at x.fn.extend.append (https://mbcomp1:9999/static/components/jquery/jquery.min.js:5:18822)
    at OutputArea._safe_append (https://mbcomp1:9999/static/notebook/js/outputarea.js:336:26)

Answer 1

当使用 mpld3.save_html(fig,"./out.html") 将图形保存到文件时，您的代码在 ipython（没有记事本）上对我有效。可能是 ipython notepad/mpld3 兼容性或 mpld3.display 的问题（这对我来说是一个错误，尽管我认为这与我的旧版本的 matplotlib 有关电脑）。

对我有用的完整代码是，

import numpy as np
import matplotlib.pyplot as plt, mpld3
import seaborn as sns
import pandas as pd

N=10
data = pd.DataFrame({"x": np.random.randn(N),
                     "y": np.random.randn(N), 
                     "size": np.random.randint(20,200, size=N),
                     "label": np.arange(N)
                     })


scatter_sns = sns.lmplot("x", "y", 
           scatter_kws={"s": data["size"]},
           robust=False, # slow if true
           data=data, size=8)
fig = plt.gcf()

tooltip = mpld3.plugins.PointLabelTooltip(fig, labels=list(data.label))
mpld3.plugins.connect(fig, tooltip)

mpld3.save_html(fig,"./out.html")

Answer 2

我认为目前没有一种简单的方法可以做到这一点。我可以通过将 tooltip 构造函数替换为以下内容来获得一些工具提示：

ax = plt.gca()
pts = ax.get_children()[3]
tooltip = mpld3.plugins.PointLabelTooltip(pts, labels=list(data.label))

不过，这仅适用于不确定区间之外的点。我认为可以扩展 seaborn 以使这些点在 zorder 中最高并将它们存储在实例中的某个地方，这样您就不需要将它们从轴子列表中拉出来。也许值得一个功能请求。

Answer 3

我能够通过在 seaborn 图上使用标准的 matplotlib 散点图和非常低的 alpha（不能使用零）来使工具提示正常工作

data_tip_points = ax.scatter(x_points, y_points, alpha=0.001)
tooltip = plugins.PointLabelTooltip(data_tip_points, labels)

这有点乱七八糟，但它的工作原理如下所示。

http://nbviewer.ipython.org/urls/bitbucket.org/jeff_mcgehee/cds_presentation_intro/raw/49cc7808ec26adebec94ffa83973bb5db13017d7/CDS%20Intro%20Presentation.ipynb

如何使 mpld3 与 seaborn 一起工作（交互式工具提示）

How to make mpld3 work with seaborn (interactive tooltips)

matplotlib

seaborn

mpld3