ipecharts brings interactive widgets based on Apache ECharts charting library to the Jupyter ecosystem. By using the Jupyter Widget protocol, ipecharts is fully compatible with other widget libraries and tools in the Jupyter ecosystem.
ipecharts.mp4
Note
pyecharts also supports using Echarts in the notebook, but they are not using Jupyter Widget like ipecharts. In this library, HTML code is injected into the notebook to render the chart.
You can try it online by clicking on this badge:
You can read the documentation following this link: https://ipecharts.readthedocs.io/
To install the extension, execute:
pip install ipechartsor with conda:
conda install -c conda-forge ipechartsipecharts widgets are generated automatically from ECharts 5.5.0. It provides two high-level widgets to create charts in notebooks: EChartsRawWidget and EChartsWidget.
EChartsRawWidget is a simple widget to render ECharts option dictionary. It is fully compatible with the JavaScript version of ECharts. Here is an example of converting the following JS example:
import * as echarts from 'echarts';
var chartDom = document.getElementById('main');
var myChart = echarts.init(chartDom);
var option;
option = {
xAxis: {
type: 'category',
boundaryGap: false,
data: ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun']
},
yAxis: {
type: 'value'
},
series: [
{
data: [820, 932, 901, 934, 1290, 1330, 1320],
type: 'line',
areaStyle: {}
}
]
};
option && myChart.setOption(option);into using EChartsRawWidget:
from ipecharts import EChartsRawWidget
option = {
'xAxis': {
'type': 'category',
'boundaryGap': False,
'data': ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun']
},
'yAxis': {
'type': 'value'
},
'series': [
{
'data': [820, 932, 901, 934, 1290, 1330, 1320],
'type': 'line',
'areaStyle': {}
}
]
}
EChartsRawWidget(option=option)
While the raw widget can render the charts correctly, it lacks the interactivity of a Jupyter widget. ipecharts provides EChartsWidget and configuration classes for nearly all available options of ECharts to correct this issue.
Here is the equivalent of the above chart but using EChartsWidget:
from ipecharts import EChartsWidget
from ipecharts.option import Option, XAxis, YAxis
from ipecharts.option.series import Line
line = Line(data=[820, 932, 901, 934, 1290, 1330, 1320], areaStyle={})
option = Option(
xAxis=XAxis(
type="category",
boundaryGap=False,
data=["Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"],
),
yAxis=YAxis(type="value"),
series=[line],
)
EChartsWidget(option=option)While it looks more verbose, the advantage is the reactivity. We can update the line data and have the chart updated automatically.
Each key in the option dictionary of ECharts has an equivalent configuration class with the same name. These classes contain traits with the same name as the corresponding ECharts option. Any change to these traits will be propagated to the top-level widget, and the chart will be updated automatically.
For instance, you can compare the scatter option of ECharts at https://echarts.apache.org/en/option.html#series-scatter.type and the equivalent Scatter class in the ipecharts documentation. The Python class is generated automatically from the ECharts option.
By using Traitlets to configure your widget, you can use EChartsWidget with other widgets in the Jupyter ecosystem. Here is an example of controlling the chart with an ipywidgets Button:
from ipecharts.option import Option, XAxis, YAxis
from ipecharts.option.series import Line
from ipywidgets.widgets import Button
line = Line(smooth=True, areaStyle={}, data=numpy.random.rand(10).tolist())
option = Option(
xAxis=XAxis(type="category"),
yAxis=YAxis(type="value"),
series=[line],
)
chart = EChartsWidget(option=option)
button = Button(description="Generate data")
def on_button_clicked(b):
data = numpy.random.rand(10).tolist()
line.data = data
button.on_click(on_button_clicked)
display(button, chart)
Both EChartsWidget and EChartsRawWidget classes can customize the Echarts init parameters, for example:
chart = EChartsWidget(
option=option, renderer="svg", width="300px", height="300px", use_dirty_rect=True
)The init parameters need to be converted to the snake case format.
Both EChartsWidget and EChartsRawWidget classes allow you to customize the style of the chart container by setting the style attribute. The style attribute accepts a dictionary where keys are CSS property names in camelCase or kebab-case (as strings), and values are the corresponding CSS values.
Example: 'backgroundColor': '#f0f0f0' or 'background-color': '#f0f0f0'
from ipecharts import EChartsWidget
from ipecharts.option import Option, XAxis, YAxis
from ipecharts.option.series import Line
# Define the data for the line series
line = Line(
data=[820, 932, 901, 934, 1290, 1330, 1320],
areaStyle={}
)
# Create the option object with xAxis, yAxis, and series
option = Option(
xAxis=XAxis(
type="category",
boundaryGap=False,
data=["Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"]
),
yAxis=YAxis(type="value"),
series=[line]
)
# Define the style for the widget
style = {
'width': '450px',
'height': '300px',
'border': '5px solid #ccc'
}
# Create the EChartsWidget with the option and style
chart = EChartsWidget(option=option, style=style)
# Display the chart
chartAfter the widget has been created and displayed, you can update its style by modifying the style attribute.
# Update the style of the chart
chart.style = {
'width': '800px',
'height': '600px',
'border': '2px solid #000'
}
# The widget will automatically update to reflect the new styles.update_style.mp4
Event-handling functions can be added to EChartsWidget and EChartsRawWidget using the same syntax as in the Javascript version:
chart = EChartsWidget(option=option)
def callback(params):
print(params)
# Add event handlers
chart.on('click', None, callback) # Listen to all click event
chart.on('click', 'series.line', callback) # Using string query
chart.on('mouseover', {'seriesIndex': 1, 'name': 'xx'}, callback) # Using object query
# Remove event handlers
chart.off('click') # Remove all handler on click event
chart.off('mouseover', callback) # Remove selected handler.