I have a figure defined like this
returns_hist = figure(plot_width=SMALLWIDTH, plot_height=HEIGHT, tools='', title='Returns Histogram')
I then update the data in an update function in this way
hist, edges = np.histogram(data['returns'], density=True, bins=50) returns_hist.quad(top=hist, bottom=0, left=edges[:-1], right=edges[1:], line_color="white")
What I see is that some of the data sets have very large central peaks. Other's small.
If I display a large peeked data set its fine, the y_axis rescales.
If I then display a data set with a low peak, the y_axis does not rescale.
Is there a way for forcing a rescaling?