Forgot "reply-to-all".
Eric
Thanks Eric!
Some followup questions.
-Can I give it more of a 3d look? Like the pie charts in MS Word?
-How can I reduce the size of the canvas behind the chart? There is a
lot of empty whitespace and I need to tighten up the margins as the
graph will be displayed inline on an html page...
-for the colors, do I use numbers to customize the colors, or do I
have to use colors from a built in pallet?
-lastly - is it possible to print the labels inside each slice of the
chart, instead of outside?
Thanks!
Erik
···
On 6/3/07, Eric Firing <efiring@...202...> wrote:
Forgot "reply-to-all".
Eric
---------- Forwarded message ----------
From: Eric Firing <efiring@...202...>
To: Erik Wickstrom <erik@...1629...>
Date: Sun, 03 Jun 2007 16:19:20 -1000
Subject: Re: [Matplotlib-users] Pie chart is "stretched"
Erik Wickstrom wrote:
> Hi all,
>
> I'm trying to generate a pie chart to use in my django web app. But
> it keeps rendering in a rectanuglar canvas instead of a square one, so
see inserted call to set_aspect method, below.
> the pie is oblong instead of circular. It also keeps rendering with a
> grey background, I need the background to be white.
see inserted call to set_facecolor method, below.
>
> I've attached a jpg of the chart - here is my source.
Do you really want jpeg? It is suitable for photos, but very inefficient
(and a bit fuzzy) for things like pie charts. I would use png format
instead, which is efficient for this sort of plot.>
> def chart(request):
> from PIL import Image as PILImage
> from matplotlib.backends.backend_agg import FigureCanvasAgg as
> FigureCanvas
> from matplotlib.figure import Figure
> from StringIO import StringIO
> fig = Figure()
> canvas = FigureCanvas(fig)
> ax = fig.add_subplot(111)
> #ax.plot([1,2,3])
> labels = 'Frogs', 'Hogs', 'Dogs', 'Logs'
> fracs = [15,30,45, 10]
> ax.pie(fracs, labels=labels, shadow=True)ax.set_aspect('equal', adjustable='box')
fig.set_facecolor('w')> #ax.set_title('hi mom')
> ax.grid(True)
> #ax.set_xlabel('time')
> #ax.set_ylabel('volts')
> canvas.draw()
> size = canvas.get_renderer().get_canvas_width_height()
> buf=canvas.tostring_rgb()
> im=PILImage.fromstring('RGB', size, buf, 'raw', 'RGB', 0, 1)
> imdata=StringIO()
> im.save(imdata, format='JPEG')
> response = HttpResponse(imdata.getvalue(), mimetype='image/jpeg')
> return response
>
> Also - can I choose my own colors for the slices?Yes, use the "colors" kwarg to specify your own list. From the docstring:
PIE(x, explode=None, labels=None,
colors=('b', 'g', 'r', 'c', 'm', 'y', 'k', 'w'),
autopct=None, pctdistance=0.6, shadow=False)Eric
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users
I didn't see Eric answer your "oblong" part of the question. To make
it circular, use aspect='equal'
ax = fig.add_subplot(111, aspect='equal')
-Can I give it more of a 3d look? Like the pie charts in MS Word?
enable shadow=True, eg
pie(...., shadow=True)
-How can I reduce the size of the canvas behind the chart? There is a
lot of empty whitespace and I need to tighten up the margins as the
graph will be displayed inline on an html page...
Set your own axes dimensions, just make sure it is square
fig = figure(figsize=(6,6))
ax = fig.add_axes([0.05, 0.05, 0.9, 0.9], aspect='equal')
-for the colors, do I use numbers to customize the colors, or do I
have to use colors from a built in pallet?
You can do it either way, but if you want your colors to reflect the
magnitude of the percentage for a given slice, you will need to use a
colormap. See below
-lastly - is it possible to print the labels inside each slice of the
chart, instead of outside?
Unfortunately not, but there should be. I just added a new keyword
arg "labeldistance" which is a fraction of the radius at which to
print the label (default 1.1). Changes are in svn.
import matplotlib.cm as cm
from pylab import figure, show
# make a square figure and axes
fig = figure(figsize=(8,8))
ax = fig.add_axes([0.05, 0.05, 0.90, 0.90], aspect='equal')
labels = 'Frogs', 'Hogs', 'Dogs', 'Logs'
fracs = [15,30,45, 10]
jet = cm.get_cmap('RdYlGn', 256)
scalarmap = cm.ScalarMappable(cmap=jet)
scalarmap.set_clim(0, 100)
colors = [scalarmap.to_rgba(frac) for frac in fracs]
ax.pie(fracs, labels=labels, colors=colors,
pctdistance=0.4, labeldistance=0.9,
autopct='%d%%', shadow=True)
show()
···
On 6/4/07, Erik Wickstrom <erik@...1629...> wrote:
Hi John,
I'm already using shadow=True - is it possible to increase the size of
the shadow?
I tried fig = figure(figsize=(6,6))
ax = fig.add_axes([0.05, 0.05, 0.9, 0.9], aspect='equal')
figsize seems to change the size of the whole figure - I want a large
chart, but less whitespace behind (tighter margins).
When I use fig.add_axes() I get a grid around my pie chart.
Also - how do I change the font used in the labels? I want to use
something like Ariel.
Thanks for your help!!
Erik
···
On 6/4/07, John Hunter <jdh2358@...287...> wrote:
On 6/4/07, Erik Wickstrom <erik@...1629...> wrote:
I didn't see Eric answer your "oblong" part of the question. To make
it circular, use aspect='equal'ax = fig.add_subplot(111, aspect='equal')
> -Can I give it more of a 3d look? Like the pie charts in MS Word?
enable shadow=True, eg
pie(...., shadow=True)
> -How can I reduce the size of the canvas behind the chart? There is a
> lot of empty whitespace and I need to tighten up the margins as the
> graph will be displayed inline on an html page...Set your own axes dimensions, just make sure it is square
fig = figure(figsize=(6,6))
ax = fig.add_axes([0.05, 0.05, 0.9, 0.9], aspect='equal')> -for the colors, do I use numbers to customize the colors, or do I
> have to use colors from a built in pallet?You can do it either way, but if you want your colors to reflect the
magnitude of the percentage for a given slice, you will need to use a
colormap. See below> -lastly - is it possible to print the labels inside each slice of the
> chart, instead of outside?Unfortunately not, but there should be. I just added a new keyword
arg "labeldistance" which is a fraction of the radius at which to
print the label (default 1.1). Changes are in svn.import matplotlib.cm as cm
from pylab import figure, show
# make a square figure and axes
fig = figure(figsize=(8,8))
ax = fig.add_axes([0.05, 0.05, 0.90, 0.90], aspect='equal')labels = 'Frogs', 'Hogs', 'Dogs', 'Logs'
fracs = [15,30,45, 10]jet = cm.get_cmap('RdYlGn', 256)
scalarmap = cm.ScalarMappable(cmap=jet)
scalarmap.set_clim(0, 100)colors = [scalarmap.to_rgba(frac) for frac in fracs]
ax.pie(fracs, labels=labels, colors=colors,
pctdistance=0.4, labeldistance=0.9,
autopct='%d%%', shadow=True)show()