I have an area plot where I would like to use colors to make it very clear when
How can I show positive values with green color and negative values with read color?
How can I use a colormap with the area chart to indicate large (absolute) values?
import numpy as np
import holoviews as hv
import panel as pn
hv.extension('bokeh')
xs = np.linspace(0, np.pi*4, 40)
area = hv.Area((xs, np.sin(xs)))
pn.panel(area).servable()
My first reaction was to try
xs = np.linspace(0, np.pi*4, 40)
ys = np.sin(xs)
hv.Area((xs, [max(v,0) for v in ys]))*hv.Area((xs, [min(v,0) for v in ys]))
but this does something interesting for y=0…
I thought something like:
https://holoviews.org/user_guide/Style_Mapping.html
import numpy as np
import holoviews as hv
import panel as pn
hv.extension('bokeh')
xs = np.linspace(0, np.pi*4, 40)
area = hv.Area((xs, np.sin(xs)))
area.apply.opts(color=hv.dim('y').bin([-2, 0, 2], ["black", "red"]))
But yields:
Mapping a dimension to the “color” style option is not supported by the Area element using the bokeh backend. To map the “y” dimension to the color use a groupby operation to overlay your data along the dimension.
So maybe a groupby
hvplot seems to work:
import numpy as np
import holoviews as hv
import panel as pn
import pandas as pd
hv.extension('bokeh')
xs = np.linspace(0, np.pi*4, 40)
ys = np.sin(xs)
zs = ys > 0
df = pd.DataFrame({"x": xs, "y": ys, "z": zs})
ds = hv.Dataset(df)
ds.to.area("x", "y", "z").overlay()
import hvplot.pandas
df.hvplot.area('x', 'y', by='z')
