Themes

Load cuxfilter and data

from cuxfilter import charts
import cuxfilter
from bokeh import palettes
cux_df = cuxfilter.DataFrame.from_arrow('../../../data/146M_predictions_v2.arrow')

Declare charts

#tile provider for choropleth map
from bokeh.tile_providers import get_provider as gp
tile_provider = gp('CARTODBPOSITRON')
tile_provider_dark = gp('STAMEN_TONER')

chart1 = charts.bokeh.choropleth(x='zip', y='delinquency_12_prediction', aggregate_fn='mean',
                                  geoJSONSource = 'https://raw.githubusercontent.com/rapidsai/cuxfilter/GTC-2018-mortgage-visualization/javascript/demos/GTC%20demo/public/data/zip3-ms-rhs-lessprops.json',
                                  data_points=1000,
                                  nan_color='white', tile_provider= tile_provider)
chart2 = charts.bokeh.bar('borrower_credit_score')
chart_widget = charts.panel_widgets.multi_select('dti')

#create a list of charts
charts_list = [chart_widget, chart1, chart2]

Themes Types

1. Light

d = cux_df.dashboard(charts_list, layout=cuxfilter.layouts.feature_and_base, theme=cuxfilter.themes.light, title="Theme - light")
await d.preview()

2. Rapids

d = cux_df.dashboard(charts_list, layout=cuxfilter.layouts.feature_and_base, theme=cuxfilter.themes.rapids, title="Theme - rapids")
await d.preview()

3. Dark

chart1.tile_provider = tile_provider_dark
d = cux_df.dashboard(charts_list, layout=cuxfilter.layouts.feature_and_base, theme=cuxfilter.themes.dark, title="Theme - dark")
await d.preview()