Nearly everyone: use https://discourse.holoviz.org/ instead! This chat is for informal discussions with developers of the HoloViz.org tools (HoloViews, Datashader, Panel, hvPlot, GeoViews, Param, etc.), but user questions should go to the HoloViz Discourse site
@jbednar
@rindis , I'd recommend not ever switching the order like that; it currently works because of a quirk in how Bokeh handles "annotations" like map tiles, but it should not be ok to switch the order of an overlay in HoloViews. So if it doesn't work when you overlay
map_tiles * plot, that's a problem that should be reported and addressed.
Didn't know that, thanks! I'll report it next time i see it :)
import holoviews as hv
import panel as pn
import geoviews.tile_sources as gts
hv.extension('bokeh')
pn.Tabs(
('tab 1', gts.tile_sources['EsriOceanBase']),
('tab 2', gts.tile_sources['EsriOceanBase'])
)I'm struggelig to make GripSpec do what I want, and it is challenging to explain! However, below is the code that has gotten me closest to what I'm trying to acheive. The blue and green spacers represent the positions where I want to put plots, the red ones are spacers added to get the behaviour I want. So, firstly I want the blue spacer to always have an aspect ratio of 1, and the green spacer should have a max height of 300. With those restrictions the browser window will be split horizontally in half. In the lower half, the green spacer will span the whole width and fill the height up to 300. For the upper half, the blue spacer should be positioned in the center, fill the whole height, and fill the width to acheive aspect ratio of 1. With the below code this works partially, where I have also posted some gifs showing the behavior. When reducing the width of the browser, at some point the blue spacer starts to reduce more in size than necessary, because the red spacers seems to take precedence below that size. When changing the browser window height, the green spacer never changes.
Any input is greatly appreciated.
import panel as pn
import geoviews.tile_sources as gts
pn.extension()
grid_layout = pn.GridSpec(sizing_mode='stretch_both')
grid_layout[0,0] = pn.Spacer(background='#FF0000')
grid_layout[0,1] = pn.Spacer(background='blue', aspect_ratio=1)
grid_layout[0,2] = pn.Spacer(background='#FF0000', sizing_mode='stretch_height')
grid_layout[1,:] = pn.Spacer(background='green', max_height=300)
grid_layout.servable()
Is it possible to have two 'unlinked' map tile plots? Im trying the following, but the plots are linked, if I zoom in on one of them, the other one also zooms in.
pn.Tabs(
('tab 1', pn.panel(gts.tile_sources['EsriOceanBase'], linked_axes=False)),
('tab 2', pn.panel(gts.tile_sources['EsriOceanBase'], linked_axes=False))
)
import panel as pn
import geoviews.tile_sources as gts
pn.extension()
grid_layout = pn.GridSpec(sizing_mode='stretch_both')
grid_layout[0,0] = pn.Spacer(min_height=0, min_width=0, background='#FF0000')
grid_layout[0,1] = pn.Spacer(background='blue', aspect_ratio=1)
grid_layout[0,2] = pn.Spacer(min_height=0, min_width=0, background='#FF0000', sizing_mode='stretch_height')
grid_layout[1,:] = pn.Spacer(background='green', max_height=300)
grid_layout.servable()
Silly remark that confused me in documentation. If I have data dict that does not define x,y that I need to plot. How do I do it? some sore of dimension / relabel function? It took me an hour to figure out I needed kdims.
Here is the code that modified the simple example from http://holoviews.org/user_guide/Style_Mapping.html
import panel
import holoviews
from holoviews import opts
from bokeh.resources import INLINE
holoviews.extension('bokeh')
data = {
'Param': [0, 1, 0.5],
'Value': [1, 0, 0.5],
'alpha': [0.5, 1, 0.3],
'color': ['red', 'blue', 'green'],
'marker': ['circle', 'triangle', 'diamond'],
'size': [15, 25, 40]
}
opts.defaults(opts.Points(padding=0.1, size=8, line_color='black'))
Plot = holoviews.Points(data, vdims=['alpha', 'color', 'marker', 'size']).opts(alpha='alpha', color='color', marker='marker', size='size')
Out=panel.panel(Plot)
Out.save('TestOut', resources=INLINE)
`This code breaks. It took me about an hour to figure out that I should use kdims. like this:
Plot = holoviews.Points(data, kdims=['Param','Value'], vdims=['alpha', 'color', 'marker', 'size']).opts(alpha='alpha', color='color', marker='marker', size='size')Do you thin its worthwhile changing the docs to use the dictionary that does not use x,y? Just to explain that those are defaults?
Its just an idea - I figured it may help newcomers.