使用带有 LinkedBrush 的 matplotlib mpld3 的 Json 序列化错误

Question

LinkedBrush 的 mpld3（matplotlib on d3）示例 http://mpld3.github.io/examples/linked_brush.html 提供以下代码示例：

import numpy as np
import matplotlib
import matplotlib.pyplot as plt
from sklearn.datasets import load_iris

import mpld3
from mpld3 import plugins, utils


data = load_iris()
X = data.data
y = data.target

# dither the data for clearer plotting
X += 0.1 * np.random.random(X.shape)

fig, ax = plt.subplots(4, 4, sharex="col", sharey="row", figsize=(8, 8))
fig.subplots_adjust(left=0.05, right=0.95, bottom=0.05, top=0.95,
                    hspace=0.1, wspace=0.1)

for i in range(4):
    for j in range(4):
        points = ax[3 - i, j].scatter(X[:, j], X[:, i],
                                      c=y, s=40, alpha=0.6)

# remove tick labels
for axi in ax.flat:
    for axis in [axi.xaxis, axi.yaxis]:
        axis.set_major_formatter(plt.NullFormatter())

# Here we connect the linked brush plugin
plugins.connect(fig, plugins.LinkedBrush(points))

mpld3.show()

虽然公共网页显示链接输出矩阵，但在本地运行时会出现json 序列化错误：

Traceback (most recent call last):
  File "/git/scalatesting/src/main/python/mpld3_linkedbrush.py", line 34, in <module>
    mpld3.show()
  File "/usr/local/lib/python2.7/site-packages/mpld3/_display.py", line 358, in show
    html = fig_to_html(fig, **kwargs)
  File "/usr/local/lib/python2.7/site-packages/mpld3/_display.py", line 251, in fig_to_html
    figure_json=json.dumps(figure_json, cls=NumpyEncoder),
  File "/usr/local/Cellar/python/2.7.14/Frameworks/Python.framework/Versions/2.7/lib/python2.7/json/__init__.py", line 251, in dumps
    sort_keys=sort_keys, **kw).encode(obj)
  File "/usr/local/Cellar/python/2.7.14/Frameworks/Python.framework/Versions/2.7/lib/python2.7/json/encoder.py", line 207, in encode
    chunks = self.iterencode(o, _one_shot=True)
  File "/usr/local/Cellar/python/2.7.14/Frameworks/Python.framework/Versions/2.7/lib/python2.7/json/encoder.py", line 270, in iterencode
    return _iterencode(o, 0)
  File "/usr/local/lib/python2.7/site-packages/mpld3/_display.py", line 138, in default
    return json.JSONEncoder.default(self, obj)
  File "/usr/local/Cellar/python/2.7.14/Frameworks/Python.framework/Versions/2.7/lib/python2.7/json/encoder.py", line 184, in default
    raise TypeError(repr(o) + " is not JSON serializable")
TypeError: array([ 1.]) is not JSON serializable

本地环境是

$pip show mpld3
Name: mpld3
Version: 0.3
Summary: D3 Viewer for Matplotlib
Home-page: http://mpld3.github.com
Author: Jake VanderPlas
Author-email: jakevdp@cs.washington.edu
License: BSD 3-clause
Location: /usr/local/lib/python2.7/site-packages

和

$python -V
Python 2.7.14

mpld3 今天安装了。是否存在mpld3 版本控制问题？还有其他建议吗？

Answer 1

根据@snakecharmerb 的评论，我从mpld3 分叉，输入了建议的修复程序，并从我在 github 上的新分支安装了 pip。

修复在这里：https://github.com/javadba/mpld3/tree/display_fix。它可以通过以下方式安装：

python -m pip install --user "git+https://github.com/javadba/mpld3@display_fix"

效果很好：json serialization 错误消失了，9 个图表之间的链接功能正常：

Answer 2

对我来说，here 给出的解决方案不起作用。

我有一个 networkx 图来可视化：

import matplotlib.pyplot as plt
import numpy as np
import mpld3

import networkx as nx
G = nx.path_graph(4)
pos = nx.spring_layout(G)

fig, ax = plt.subplots(subplot_kw=dict(facecolor='#EEEEEE'))
scatter = nx.draw_networkx_nodes(G, pos, ax=ax)
nx.draw_networkx_edges(G, pos, ax=ax)

labels = G.nodes()
tooltip = mpld3.plugins.PointLabelTooltip(scatter, labels=labels)
mpld3.plugins.connect(fig, tooltip)

mpld3.show()

然后它给出了“JSON 不可序列化”错误。我找到了上面的链接，并尝试了修复。修复本质上说如果对象是numpy.ndarray 类型，则将其更改为列表。

但是G.nodes的对象类型是networkx.classes.reportviews.NodeView，而不是numpy.ndarray；因此它不起作用。

因此，我修改了文件_display.py 以添加import networkx 并在NumpyEncoder 类的default 函数中添加以下两行以使其工作：

elif isinstance(obj,networkx.classes.reportviews.NodeView):
    return list(obj)

现在可以了：