Zane Selvans wrote: > Hey there, > > I'm trying to plot a bunch-o-vectors, colored red or blue, depending on > whether their magnitude is positive or negative (they represent > stresses), and I'm doing something like this: > > mag1 = evals[:,0] > ex1 = evecs[:,0,1] > ey1 = evecs[:,0,0] > C1 = np.where(mag1 >= 0, 'red', 'blue') > > mag2 = evals[:,1] > ex2 = evecs[:,1,1] > ey2 = evecs[:,1,0] > C2 = np.where(mag2 >= 0, 'red', 'blue') >
The C argument to quiver is for an array to be used with a colormap; you could use that with a listed colormap, or, to use something very close to what you already have, try using the "color" kwarg for your C1 and C2. It takes a single value or a sequence of color specs. Eric > basemap_ax.quiver(np.degrees(calc_phis), > np.degrees(np.pi/2.0-calc_thetas), mag1*ex1, mag1*ey1, C1, lw=0, > width=0.002, scale=1e8) > basemap_ax.quiver(np.degrees(calc_phis), > np.degrees(np.pi/2.0-calc_thetas), mag2*ex2, mag2*ey2, C2, lw=0, > width=0.002, scale=1e8) > basemap_ax.quiver(np.degrees(calc_phis), > np.degrees(np.pi/2.0-calc_thetas), -mag1*ex1, -mag1*ey1, C1, lw=0, > width=0.002, scale=1e8) > basemap_ax.quiver(np.degrees(calc_phis), > np.degrees(np.pi/2.0-calc_thetas), -mag2*ex2, -mag2*ey2, C2, lw=0, > width=0.002, scale=1e8) > > And it gives me the stack trace below... I also tried just using > plain-old quiver (not via basemap) and got the same error. I tried > using rgba tuples and grayscale strings as colors as well, and still got > the same error. Anybody have any idea what the deal is? Or am I > misunderstanding what *C* is supposed to be (i.e. not just an array of > colors, of the same length as the number of vectors being plotted...) > > /Library/Python/2.5/site-packages/mpl_toolkits/basemap/__init__.pyc in > quiver(self, x, y, u, v, *args, **kwargs) > 2877 ax.hold(h) > 2878 try: > -> 2879 ret = ax.quiver(x,y,u,v,*args,**kwargs) > 2880 try: > 2881 plt.draw_if_interactive() > > /Library/Python/2.5/site-packages/matplotlib/axes.pyc in quiver(self, > *args, **kw) > 5850 def quiver(self, *args, **kw): > 5851 if not self._hold: self.cla() > -> 5852 q = mquiver.Quiver(self, *args, **kw) > 5853 self.add_collection(q, False) > 5854 self.update_datalim(q.XY) > > /Library/Python/2.5/site-packages/matplotlib/quiver.pyc in > __init__(self, ax, *args, **kw) > 367 **kw) > 368 self.polykw = kw > --> 369 self.set_UVC(U, V, C) > 370 self._initialized = False > 371 > > /Library/Python/2.5/site-packages/matplotlib/quiver.pyc in set_UVC(self, > U, V, C) > 439 mask = ma.mask_or(U.mask, V.mask, copy=False, shrink=True) > 440 if C is not None: > --> 441 C = ma.masked_invalid(C, copy=False).ravel() > 442 mask = ma.mask_or(mask, C.mask, copy=False, shrink=True) > 443 if mask is ma.nomask: > > /Library/Python/2.5/site-packages/numpy/ma/core.pyc in masked_invalid(a, > copy) > 1996 """ > 1997 a = np.array(a, copy=copy, subok=True) > -> 1998 condition = ~(np.isfinite(a)) > 1999 if hasattr(a, '_mask'): > 2000 condition = mask_or(condition, a._mask) > > TypeError: bad operand type for unary ~: 'NotImplementedType' > > > -- > Zane A. Selvans > Amateur Earthling > http://zaneselvans.org > +1 303 815 6866 > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------------ > Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day > trial. Simplify your report design, integration and deployment - and focus on > what you do best, core application coding. Discover what's new with > Crystal Reports now. http://p.sf.net/sfu/bobj-july > > > ------------------------------------------------------------------------ > > _______________________________________________ > Matplotlib-users mailing list > Matplotlib-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/matplotlib-users ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ Matplotlib-users mailing list Matplotlib-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-users
