matplotlib-users Mailing List for matplotlib

Hi, im trying to create a heatmap/colourmap/contour style plot similar to
http://www.idlcoyote.com/cg_tips/outcontourbar.png but I cant seem to get it
working correctly. The code takes a number of parameters on the command line
because it is passed by an external PHP script. Essentially its a list of
lat/lon/values which I want to interpolate and plot but I cant get anything
like the image above, I cant get the vmin/vmax to work for the colorbar and
I dont know why the whole plot is squared off rather than fading to white if
you see what I mean
import os 
import tempfile
import math
os.environ['MPLCONFIGDIR'] = tempfile.mkdtemp()
import argparse
import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import scipy.interpolate
width = 800
height = 600
lat_min = []
lon_min = []
lat_max = []
lon_max = []
# assumes lat is y, lon is x, and image is 800x600
def latToXY(lat):
	global width, height, lat_min, lat_max
	y = ((lat - lat_min) / (lat_max - lat_min)) * height
	#print y
	return y
def lonToXY(lon):
	global width, height, lon_min, lon_max
	lon = math.fabs(lon)	
	x = ((lon - lon_min) / (lon_max - lon_min)) * width
	#print x
	return x
def scipy_idw(x, y, z, xi, yi):
 #interp = scipy.interpolate.Rbf(x, y, z, function='linear')
 interp = scipy.interpolate.Rbf(x, y, z)
 return interp(xi, yi)
def plot(x,y,z,grid,legend_min,legend_max,filename):
 plt.figure()
 fig = plt.imshow(grid, vmin=legend_min, vmax=legend_max, extent=[0,
1024, 0, 768])
 fig.axes.get_xaxis().set_visible(False) # hide axis labels
 fig.axes.get_yaxis().set_visible(False)
 #plt.hold(True)
 plt.scatter(x,y,c=z)
 plt.colorbar()
 plt.savefig(filename)
# grab all floats from command line
parser = argparse.ArgumentParser()
parser.add_argument('--l1', type=str)
parser.add_argument('--l2', type=str)
parser.add_argument('--l3', type=str)
parser.add_argument('--min', type=str)
parser.add_argument('--max', type=str)
parser.add_argument('--filename', type=str)
args = parser.parse_args()
# create a list by splitting at the comma
l1_list = args.l1.split(',') # ['1','2','3','4']
l2_list = args.l2.split(',')
l3_list = args.l3.split(',')
legend_min = float(args.min)
legend_max = float(args.max)
filename = args.filename
# convert string list to list of floats
for i in range(len(l1_list)):
	l1_list[i] = float(l1_list[i]) 
	l2_list[i] = float(l2_list[i]) 
	l3_list[i] = float(l3_list[i])
	
lat_min = min( math.fabs(yy) for yy in l2_list )
lat_max = max( math.fabs(yy) for yy in l2_list )
lon_min = min( math.fabs(xx) for xx in l1_list )
lon_max = max( math.fabs(xx) for xx in l1_list )
# convert list of floats to x,y
for i in range(len(l1_list)):
	l1_list[i] = lonToXY(l1_list[i])
	l2_list[i] = latToXY(l2_list[i])
# convert list to numpy array
x = np.array(l1_list)
y = np.array(l2_list)
z = np.array(l3_list)
#print x
#print y
#print z
nx, ny = 50, 50
xi, yi = np.linspace(x.min(), x.max(), nx), np.linspace(y.min(), y.max(),
ny)
xi, yi = np.meshgrid(xi, yi)
xi, yi = xi.flatten(), yi.flatten()
grid2 = scipy_idw(x,y,z,xi,yi)
grid2 = grid2.reshape((ny, nx))
plot(x,y,z,grid2,legend_min,legend_max,filename)
I call the script with the following parameters:
plot.py --l1=-1.8791363,-1.8786206,-1.8796862,-1.878171
--l2=57.458459,57.458153,57.458495,57.458036 --l3=42.3,37.8,43.5,47.7
--min=0 --max=100 --filename=/tmp/plot.png
Any help is greatly appreciated
Thanks 
--
View this message in context: http://matplotlib.1069221.n5.nabble.com/matplolib-imshow-contour-heatmap-tp43078.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

I have used Anaconda with my students because it installs a standard 
environment on all platforms, it works very well and is easy to install.
I have also tried to Enthought Canopy but swicthed to Anaconda because 
Anaconda was as simple to use, came with hfewer restrictions and in my 
experience also fewer bugs and problems than Canopy.
Macorts also work very well, though. So that is a matter of taste, I 
guess. The upside to Anaconda is that I believe it runs the Qt4Agg 
backend by default for all platforms. Can anyone confirm this (I don't 
have access to a Mac at the moment)?
On Thu 13 Mar 2014 09:29:38 PM CET, Sterling Smith wrote:
> +1 for macports
> (I haven't used the others.)
>
> On Mar 13, 2014, at 10:12AM, Felix Patzelt wrote:
>
>> Are you sure that you want to use Python 3.3 on OSX 10.6??? Do you really still use 10.6? Do you want Python 3? I'm not sure on the current status, but many projects took quite a while to get ported over from Python 2. Furthermore, as often with free software, installation can be a bit tricky. It is certainly a very different experience than installing "normal" Mac applications.
>>
>> For a bit of context, most Linux distribution have some version of Python / Matplotlib in their respective package managers. These are easily installed if the particular package manager on your Linux offers the versions you want. Otherwise, you will have to do some work.
>>
>> OSX does not have an official package manager, but there are several inofficial options. I'm using http://www.macports.org which is slow because it installs its own private versions for everything, but it works very well. This is probably the easiest way to get all the open source stuff you want on your Mac and I use it a lot. Another popular and more lightweight package manager is homebrew, which relies more on the system libraries from Apple.
>>
>> The minimal installation instructions without a package manager seem to be these: https://github.com/rueckstiess/mtools/wiki/matplotlib-Installation-Guide If you're a real unix hacker, you can install everything from source. I did that before, and it takes a lot of time and in-depth knowledge.
>>
>> Finally, there are several pre-packaged distributions like https://www.enthought.com or https://store.continuum.io/cshop/anaconda/ (see http://penandpants.com/install-python/). They might come with a normal OSX installer. Maybe https://code.google.com/p/spyderlib/ does the trick for you?
>>
>> Anyway, these are just some suggestions. Maybe you want to start a separate thread on the mailing list about the best way to install matplotlib on a mac. Please note that I cannot comment in detail on any of the installation methods that I didn't use myself.
>>
>>
>>
>>
>> Am 13.03.2014 um 17:36 schrieb Christophe Bal <pro...@gm...>:
>>
>>> I've tested a more simpler Python code.
>>>
>>> from pylab import *
>>> plot([1,2,3])
>>> show()
>>>
>>> This gives me a scary backend MacOSX version unknown. I've used the official DMG installer matplotlib-1.3.1-py3.3-python.org-macosx10.6.dmg.
>>>
>>> This seems to be a big problem. No ?
>>>
>>>
>>> $HOME=/Users/xxxx
>>> matplotlib data path /Library/Frameworks/Python.framework/Versions/3.3/lib/python3.3/site-packages/matplotlib/mpl-data
>>> loaded rc file /Library/Frameworks/Python.framework/Versions/3.3/lib/python3.3/site-packages/matplotlib/mpl-data/matplotlibrc
>>> matplotlib version 1.3.1
>>> verbose.level helpful
>>> interactive is False
>>> platform is darwin
>>> CACHEDIR=/Users/xxxx/.matplotlib
>>> Using fontManager instance from /Users/xxxx/.matplotlib/fontList.py3k.cache
>>> backend MacOSX version unknown
>>>
>>>
>>> 2014年03月13日 17:31 GMT+01:00 Felix Patzelt <fe...@ne...>:
>>> Well, there is a list in ~/.matplotlib/matplotlibrc (see http://matplotlib.org/users/customizing.html)
>>>
>>>> #### CONFIGURATION BEGINS HERE
>>>>
>>>> # the default backend; one of GTK GTKAgg GTKCairo CocoaAgg FltkAgg
>>>> # MacOSX QtAgg Qt4Agg TkAgg WX WXAgg Agg Cairo GDK PS PDF SVG Template
>>>> # You can also deploy your own backend outside of matplotlib by
>>>> # referring to the module name (which must be in the PYTHONPATH) as
>>>> # 'module://my_backend'
>>>> backend : Qt4Agg
>>>
>>> see also: http://stackoverflow.com/questions/5091993/list-of-all-available-matplotlib-backends
>>>
>>> I'm not sure about the dependencies, I guess you have to check out each one of them. If you don't use a package manager, resolving all dependency issues might be quite painful.
>>>
>>> Best,
>>> Felix Patzelt
>>>
>>> Am 13.03.2014 um 17:18 schrieb Christophe Bal <pro...@gm...>:
>>>
>>>> Thanks a lot for this big hint but neither TkAgg works nor Qt4Agg can work (because I do not have PyQt).
>>>>
>>>> Is there a complete list of all the backends ?
>>>>
>>>> Christophe BAL
>>>>
>>>
>>>
>>
>> ------------------------------------------------------------------------------
>> Learn Graph Databases - Download FREE O'Reilly Book
>> "Graph Databases" is the definitive new guide to graph databases and their
>> applications. Written by three acclaimed leaders in the field,
>> this first edition is now available. Download your free book today!
>> http://p.sf.net/sfu/13534_NeoTech_______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
> ------------------------------------------------------------------------------
> Learn Graph Databases - Download FREE O'Reilly Book
> "Graph Databases" is the definitive new guide to graph databases and their
> applications. Written by three acclaimed leaders in the field,
> this first edition is now available. Download your free book today!
> http://p.sf.net/sfu/13534_NeoTech
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users