I’m very interested in using the parallel coordinates plot in plotly.py. Given the nature of my problems, however, I usually have discrete variables for all my labels (columns), thus the lines would always be rendered on top of each other and it would be difficult to distinguish among them. It seems that the color of a segment between any two “nodes” shown in the plot would always come from the latest line plotted (assuming lines are plotted in a certain order given by the data passed in), and any previously plotted lines going through the same path would be overwritten. For example, in the plot attached (generated using plotly.graph_objs.Parcoords), there are 5 lines connecting A=0.8 and B=1, with a single color of blue shown for that segment, while none of the lines connecting B=1 and the C nodes is blue.
I was wondering if there are handlers already built-in that take care of this scenario (for example, by showing the color of the average values of the overlapped segments)? I’d also appreciate it if anyone could shed some light on how to go about the problem using the Python API when all variables are discrete.