pyecharts support show charts and export to some file formats in the Jupyter Notebook.
When install the pyecharts package using the following command, a jupyter nbextension named echarts/main will be installed if jupyter exists.Or the nbextension installation will be skipped.
pip install pyecharts
In the development, you can also use the command to install manually jupyter nbextension.
$ pip install jupyter-echarts-pypkg
You can check the jupyter nbextension using the list command.
$ jupyter nbextension list Known nbextensions: config dir: /Users/jaska/.jupyter/nbconfig notebook section echarts/main enabled - Validating: OK
In the Notebook cell ,you can simply call the instance itself to diplay the chart.
All chart classes in pyecharts implement the
_repr_html_ interface about IPython Rich Display .
Export html function
You can use Notebook's default "download as" to export other forms of files, such as ipynb or images.
NOTE: Since the exported file is out of the original Jupyter Notebook environment, in order to be able to display the chart completely, you should use the remote jshost library.
from pyecharts import online online(host='https://my-site.com') # use remote jshost
If the exported HTML file needs to display the image without a network, consider the introduction below.
Export PDF function
Users can use pyecharts-snapshot to generate static images and then output them. There are two steps that need to be performed.
- Install pyecharts-snapshot 0.1.4+ 和 phanomjs-prebuilt
- Add a statement at the beginning of the notebook. This will output the image as a PDF.
from pyecharts import Bar, configure configure(output_image='pdf') bar = Bar("我的第一个图表", "这里是副标题") bar.add("服装", ["衬衫", "羊毛衫", "雪纺衫", "裤子", "高跟鞋", "袜子"], [5, 20, 36, 10, 75, 90]) bar # Display as a pdf file in jupyter and not as a local pdf file.
Please see this example: test.pdf
Show charts in HTML offline
Users need to install pyecharts-snapshot 0.1.4+ and use the
renderer parameter to generate svg images by pyecharts 0.4.2+.
from pyecharts import configure configure(output_image='svg')
svg is also possible to output a PDF. But the output image not only requires Inkscape but the image is not very good. Therefore, it is not recommended to use
output_image='svg' to output as PDF.
Refer to pyecharts示例.
Since pyecharts 0.5.2+, users configure the theme of a single chart through the
use_theme function. If you want to configure the theme of all the charts in your environment, you can add the following statement at the very beginning of the notebook:
from pyecharts import configure configure(global_theme='dark')
Then re-run, you can see that the theme has been replaced.
From pyecharts 0.5.5+, nteract can also use pyecharts. For now, nteract users must declare
enable_nteract() at the very beginning. Otherwise, just call
enable_nteract() when you need to generate html(js) output.
However, if you need to output an image, the method is the same as the jupyter notebook mentioned above.
jupyterlab is the next generation for Jupyter Notebook,and this is a very early preview, and is not suitable for general usage yet. We will pay Continuous attention to the development and make adapter with pyecharts.