How to display Chinese characters correctly in pandas profiling?

[hengzhe-zhang]

trafficstars

I find that when I render non-ASCII characters, pandas profiling will not render them correctly. I have tried to modify the default rendering font of matplotlib. However, even though I can manually render a correct figure by using matplotlib, the figure rendered by the pandas profiler is still wrong. How can I solve this problem?

[sbrugman]

Hello @zhenlingcn,

Matplotlib unfortunately doesn't support Chinese characters by default. This can be overcome by changing the font.

See: https://jdhao.github.io/2017/05/13/guide-on-how-to-use-chinese-with-matplotlib/ https://medium.com/@hoishing/using-chinese-characters-in-matplotlib-5c49dbb6a2f7

If anyone is interested in adding this as a feature/config param, then a pull request is welcome.

Same as: https://github.com/pandas-profiling/pandas-profiling/issues/572

[hengzhe-zhang]

I have configured matplotlib correctly. However, it seems like pandas profiling does not use the default configuration of matplotlib.

[sbrugman]

@zhenlingcn Could you share that config? Then we'll try to find a solution.

[sbrugman]

Update: pandas-profiling v3 will use Altair which is not expected to have character issues.

[Blanket58]

Update: pandas-profiling v3 will use Altair which is not expected to have character issues.

With all due respect, when will the version 3 be released?

[lk137095576]

I meet the same question, I have changed the config and it show right if only use matplotlib, but pandas profiling doesn't work @sbrugman

[gsy44355]

我在我的服务器上解决了这个问题，不过方法有点蠢，只适用于硬性修改字体，不适用想配置字体的同学。 遇到这个问题的同学可以参考一下。只需要修改python包管理的中的pandas_profiling/visualisation下的plot.py 文件和missing.py `文件，在每一个绘图的function内部开头，都添加两行代码

    mpl.rcParams['font.sans-serif'] = ['SimHei']   #显示中文
    mpl.rcParams['axes.unicode_minus']=False       #显示负号

在文件开头导入包from matplotlib.pylab import mpl 就能够解决这个问题，画图就可以正常输出中文。 办法比较蠢- -如果有更好的方式请告诉我= = @lk137095576

---

English Version：

I solved this problem on my server, but the method is a bit silly and only applies to hard changes to fonts, not to guys who want to configure fonts. Just modify the plot.py file and the missing.py ` file under pandas_profiling/visualisation in the python package manager, and add two lines of code to the beginning of each function inside the plot

    mpl.rcParams['font.sans-serif'] = ['SimHei'] # show Chinese
    mpl.rcParams['axes.unicode_minus'] = False # show negative signs

Import the package from matplotlib.pylab import mpl at the beginning of the file This will solve the problem and the drawing will output Chinese properly. It's a stupid solution - - if there is a better way please let me know , thks

[cloudy-sfu]

Is there an update? I cannot implement this solution if using remote (cloud) resources like Google Colab.

我在我的服务器上解决了这个问题，不过方法有点蠢，只适用于硬性修改字体，不适用想配置字体的同学。 遇到这个问题的同学可以参考一下。只需要修改python包管理的中的pandas_profiling/visualisation下的plot.py 文件和missing.py `文件，在每一个绘图的function内部开头，都添加两行代码

    mpl.rcParams['font.sans-serif'] = ['SimHei']   #显示中文
    mpl.rcParams['axes.unicode_minus']=False       #显示负号

在文件开头导入包from matplotlib.pylab import mpl 就能够解决这个问题，画图就可以正常输出中文。 办法比较蠢- -如果有更好的方式请告诉我= = @lk137095576

English Version：

I solved this problem on my server, but the method is a bit silly and only applies to hard changes to fonts, not to guys who want to configure fonts. Just modify the plot.py file and the missing.py ` file under pandas_profiling/visualisation in the python package manager, and add two lines of code to the beginning of each function inside the plot

    mpl.rcParams['font.sans-serif'] = ['SimHei'] # show Chinese
    mpl.rcParams['axes.unicode_minus'] = False # show negative signs

Import the package from matplotlib.pylab import mpl at the beginning of the file This will solve the problem and the drawing will output Chinese properly. It's a stupid solution - - if there is a better way please let me know , thks

[JonyJiang123]

我在我的服务器上解决了这个问题，不过方法有点蠢，只适用于硬性修改字体，不适用想配置字体的同学。 遇到这个问题的同学可以参考一下。只需要修改python包管理的中的pandas_profiling/visualisation下的plot.py 文件和missing.py `文件，在每一个绘图的function内部开头，都添加两行代码

    mpl.rcParams['font.sans-serif'] = ['SimHei']   #显示中文
    mpl.rcParams['axes.unicode_minus']=False       #显示负号

在文件开头导入包from matplotlib.pylab import mpl 就能够解决这个问题，画图就可以正常输出中文。 办法比较蠢- -如果有更好的方式请告诉我= = @lk137095576

English Version：

I solved this problem on my server, but the method is a bit silly and only applies to hard changes to fonts, not to guys who want to configure fonts. Just modify the plot.py file and the missing.py ` file under pandas_profiling/visualisation in the python package manager, and add two lines of code to the beginning of each function inside the plot

    mpl.rcParams['font.sans-serif'] = ['SimHei'] # show Chinese
    mpl.rcParams['axes.unicode_minus'] = False # show negative signs

Import the package from matplotlib.pylab import mpl at the beginning of the file This will solve the problem and the drawing will output Chinese properly. It's a stupid solution - - if there is a better way please let me know , thks

it does not work for me

[whxue0807]

Add a font_path parameter in plot.py , that's ok for me : wordcloud = WordCloud( background_color="white", random_state=123, width=300, height=200, scale=2, font_path='path_to_chinese_font.ttf' # Replace 'path_to_chinese_font.ttf' with the path to a font supporting Chinese characters ).generate_from_frequencies(word_dict)