DynamicMap cannot be displayed without explicit indexing

HoloViews
, ,
1

I have created a function that has one argument: the column name of a DataFrame. According to that value, it shows a graph that has the x-axis based on that column.

def make_graphs(filter):
...
return chart

Now I want to create a dynamic map to select the column from a list and change that graph. So I created another function for that:

def show_graph():
    filter_data = ["Gender", "Seniority"]

    dyn_map = hv.DynamicMap(make_graphs, kdims=["filter"])
    dyn_map.redim.values(filter=filter_data)
    dyn_map.redim.default(filter_data[0])
    dyn_map.opts(framewise=True)
    figure = hv.render(dyn_map)

    layout = bh.layouts.row(figure, sizing_mode="scale_width")
    bh.io.show(layout)

but when I run it, I get the error message:

DynamicMap cannot be displayed without explicit indexing as 'filter' dimension(s) are unbounded. 
Set dimensions bounds with the DynamicMap redim.range or redim.values methods.
Process finished with exit code 0

Any idea why?

2

I ran into the same issue. The way I get around it is to explicitly define the dimensions rather than depending on the text parsing of Holoviews.

So in your code I’d change it to:

def show_graph():
    filter_data = ["Gender", "Seniority"]

    filter = hv.Dimension('filter', values=filter_data, default=filter_data[0])
    dyn_map = hv.DynamicMap(make_graphs, kdims=[filter]) # note removal of quotes
    dyn_map.opts(framewise=True)
    figure = hv.render(dyn_map)

    layout = bh.layouts.row(figure, sizing_mode="scale_width")
    bh.io.show(layout)

Not sure if it is a bug or not. It seems like a bug, since the examples all use redim.

3

HI. I used your code and the result is the plot for default filter, that is, “Gender”, but there is no way to Select “Seniority”. No selection widget appear. Any idea why?

4

I’ve had trouble making Bokeh’s “show” method work with dynamic plots. Instead try:

import panel as pn
pn.serve(dyn_map)

I did this with the above code and it worked.

1 Like