I had what I thought was a very simple problem but it has sent me through a rabbit hole and I guess I’m missing something. I was trying to make a very simple 2x1 plot. The first plot was a stacked bar plot, and the second one is a regular bar plot with a scatter, like this:
In this case it isn’t really a problem, but if I wanted to use two different barmodes for my subplots, what could I do? Because
barmode is linked to the figure’s
layout I can’t seem to find a way to plot two types of
barmodes in different subplots.
My code goes something like this:
trace1 = go.Bar(x=dfv['date'], y=dfv['used'], marker_color='red') trace2 = go.Bar(x=dfv.fecha, y=dfv.disponible marker_color='green') fig.append_trace(trace1, 1, 1) fig.append_trace(trace2, 1, 1) trace3 = go.Bar(x=dfv.date, y=dfv.total, name='Total ') trace4 = go.Scatter(x=dfv['date'], y=dfv['percentage'], name='% ') fig.append_trace(trace3, 2, 1) fig.append_trace(trace4, 2, 1) fig.update_layout(barmode='stack')
trace4 was also a bar and I wanted non-stacked bars for this one, what could I do?
Edit: I saw
px.bar() has a
barmode argument, but it has weird interactions with
add_trace() (or I’m also missing something, so that kinda worked, but the label was lost and I had to updated several properties with
udpate_traces. I also tried
barmode was ignored by Dash.