Open cschaerfe opened 7 years ago
If you call the plot function and store it in a variable, you can get access to the figure and axes.
upset = pyu.plot(df_dict)
You can see what 'items' are returned with upset.items()
- its a dictionary
You can 'get' any of these with upset.get()
, i.e. `upset.get('figure') will return the figure.
fig = pyu.plot(df_dict).get('figure')
The axes (4 of them in a normal plot) can be called with
ax = fig.get_axes()
Though note this returns a list...
You could alternatively get each axes with
fig = pyu.plot(df_dict).get(name)
where name is one of 'intersection_bars', 'intersection_matrix', 'base_set_size', or 'names'. Additional plots are 'additional' if you have them.
I hope that helps? It could be much more straightforward (like seaborn), I agree.
I was wondering if there is a way of passing the axis or figure the plot should be created in - similarly to how seaborn handles it? I have not seen anything like this in the code, but was wondering if I missed something - specifically for providing the figuresize etc?