creating a multi colored graph with respect to the values in y-axis

[Wanderer]

On Jun 15, 2:10 pm, Ravikanth <vvnrk.vanapa... at gmail.com> wrote:
> On Jun 15, 12:59 pm, Wanderer <wande... at dialup4less.com> wrote:
>
>
>
>
>
>
>
>
>
> > On Jun 15, 1:28 pm, Ravikanth <vvnrk.vanapa... at gmail.com> wrote:
>
> > > On Jun 15, 11:57 am, Wanderer <wande... at dialup4less.com> wrote:
>
> > > > On Jun 15, 12:00 pm, Ravikanth <vvnrk.vanapa... at gmail.com> wrote:
>
> > > > > On Jun 15, 10:32 am, Wanderer <wande... at dialup4less.com> wrote:
>
> > > > > > On Jun 15, 11:04 am, Ravikanth <vvnrk.vanapa... at gmail.com> wrote:
>
> > > > > > > Hi all,
>
> > > > > > > I am a beginner in python. I need to implement a graph with multiple
> > > > > > > colors in it.
> > > > > > > In a way, I have a function which varies with respect to time and
> > > > > > > amplitude. I have time on x-axis and amplitude on y-axis. Lets say the
> > > > > > > amplitude of the graph is divided into 4 ranges, say 1-3,3-5,5-9,
> > > > > > > 10-3. I need to plot the graph in such a way that, when the values of
> > > > > > > amplitude are in a particular range say 1-3, the color of graph should
> > > > > > > be red.
> > > > > > > If the amplitude is in the range from 3-5 the graph need to be in
> > > > > > > color blue etc..,
>
> > > > > > > Can somebody guide me on this, how to achive this functionality.
>
> > > > > > > Regards,
> > > > > > > Ravikanth
>
> > > > > > Check out the examples in matplotlib.
>
> > > > > >http://matplotlib.sourceforge.net/examples/pylab_examples/multicolore...
>
> > > > > > - Show quoted text -
>
> > > > > I did go through the side wanderer.
>
> > > > > I wasn't able to figure out the usage of boundaryNorm and
> > > > > lc.set_array(z) , in that link.
> > > > > according to my understanding,
>
> > > > > cmap = ListedColormap(['r', 'g', 'b'])
> > > > > norm = BoundaryNorm([-1, -0.5, 0.5, 1], cmap.N)
>
> > > > > In the above lines of code, as per my understanding,
> > > > > Listedcolor map, maps the colors r,g and b to specific indexes into
> > > > > cmap
> > > > > i.e cmap(0) represents red,
> > > > > cmap(1) represents blue
> > > > > cmap(2) represents green.
> > > > > for any index greater than 3 a color of blue is returned..
>
> > > > > >>> cmap = ListedColormap(['r', 'g', 'b'])
> > > > > >>> cmap(0)
>
> > > > > (1.0, 0.0, 0.0, 1.0)>>> cmap(1)
>
> > > > > (0.0, 0.5, 0.0, 1.0)>>> cmap(2)
>
> > > > > (0.0, 0.0, 1.0, 1.0)>>> cmap(3)
>
> > > > > (0.0, 0.0, 1.0, 1.0)
>
> > > > > In this context, I was not able to understand what does boundaryNorm
> > > > > does.
> > > > > We have 3 colors and we are using 4 values as argument in boundaryNorm.
> > > > > [-1, -0.5, 0.5, 1], the comment reads slope of 'z' is being mapped to
> > > > > the values in boundary norm. How is it handled.
> > > > > Does the function call " lc.set_array(z) " does it ?  what is the
> > > > > exact use of linecollection.set_array(z) in this context.
>
> > > > > Thanks,
> > > > > Ravikanth
>
> > > > The colors are referring to the slope of the line. Change
> > > > 'lc.set_array(z)' to 'lc.set_array(y)' and it might be easier to
> > > > understand.  Here are the steps.
>
> > > > 1. Define the functions x,y and z,
> > > > 2. Define the colors 'red', 'green' and 'blue' with ListedColorMap
> > > > 3. Define the three regions, (-1.0 to -0.50, -0.50 to 0.50, 0.50 to
> > > > 1.0) with BoundaryNorm([-1,-0.50, 0.50,1], cmap.N).
> > > >    (Why they add the trailing zero in 0.50 and don't change 1 to 1.0;
> > > > I don't know)
> > > > 4. Create an array of (x,y) points.
> > > > 5. Create a collection of tiny segments [(x1,y1),(x2,y2)] and color
> > > > them with cmap using the boundary rules of norm. lc =
> > > > LineCollection(segments, cmap=cmap, norm=norm)
> > > > 6. Use lc.set_array(y) to determine how to color the segments.
> > > > 7. Plot it.- Hide quoted text -
>
> > > > - Show quoted text -
>
> > > Hi Wanderer,
>
> > > Thanks for your help. It works now.
> > > Just wanted to know how to go about when I have to do my color mapping
> > > not only with respect to range of values on y-axis but also based on
> > > some other conditions as well. i.e, say ( range && <condtion1> &&
> > > <condition2> ) where condition1 could be occurance of some event say,
> > > a flag1 is set true and condition2 may be another flag2 set to false.
> > > Just wanted to use my color mapping not only based on boundaries but
> > > also on occurance of other events as well.
> > > In this context do i have to modify the source of BoundaryNorm in
> > > matplotlib function...?? Can you give me some insights into this.
>
> > > Regards,
> > > Ravikanth
>
> > I don't know if there is another way, but I think changing the
> > lc.set_array input would be the easiest. Each point has an (x,y,z)
> > where z determines the color by what range it is in. You would use
> > your conditions to set points in z to the desired color.- Hide quoted text -
>
> > - Show quoted text -
>
> does (x,y,z) mean adding another dimension 'z' (which is to be used as
> a condition) to the step 4) as suggested in the steps outlined to me
> before ?

In the matplotlib example they x,y and z.

x = np.linspace(0, 3 * np.pi, 500)
y = np.sin(x)
z = np.cos(0.5 * (x[:-1] + x[1:]))  # first derivative

Where x and y are the coordinates of the plot and z is the constraint
used to color the line. You can modify z with you conditions by adding
code that looks like

for i, xvalue in enumerate(x);
    if mycondition(xvalue):
        z[i] = 0

lc.set_array(z)

So for values of x that meet your condition, they get the color for 0.