an attempt to allow creating interactive plots from python in jupyter using echarts although it worked fine when it was tested with a lot of data points it was very slow to render R gg plot was less than a second and echarts was 5-6 mins if i remember correctly a few million data points were used this approach is still 100% doable but only for small/medium size data
UPlot(df).sankey(
'month','origin', 'distance',
add_filters=True
).size(
width=500, height=500
).render()
UPlot(df).time_series(
'iso_date', ['dep_delay', 'arr_delay']
).slider(
"x", start_pct=30, end_pct=60,
).tooltip().size(width=1000, height=500).render()
UPlot(df).calendar_heatmap(
'iso_date', 'cnt', filter_type='piecewise'
).size(width=800, height=300).render()
UPlot(df).scatter_plot(
"distance", "avg_speed", symbol_size=2
).slider('x', start_pct=0, end_pct=30).size(width=500, height=500).render()