python - Matplotlib: error with "height" in grouped barchart -

i want produce barchart 2 categories on x axis and, each category, 5 different series. took inspiration here, , amended code follows:

import numpy np  import matplotlib.pyplot plt pylab import * %pylab inline  myarray=np.array([['series1', 'a',45],            ['series2', 'a',47],            ['series3', 'a',48],            ['series4','a',48],            ['series5', 'a',49],            ['series6','b',39],            ['series7','b',37],            ['series8','b',38],                 ['series9','b',36],               ['series10','b',38]])   fig1=plt.figure() ax1=fig1.add_subplot(111) space=0.25  slots=np.unique(myarray[:,0]) categories=np.unique(myarray[:,1])  n=len(slots) width = (1 - space) / (len(slots))  i,cond in enumerate(slots):     print "cond:", cond     vals = myarray[myarray[:,0] == cond][:,2]     pos = [j - (1 - space) / 2. + * width j in range(1,len(categories)+1)]     ax1.bar(pos, vals, width=width,label=cond,color=cm.accent(float(i)/n)) 

i keep getting same error: valueerror: incompatible sizes: argument 'height' must length 2 or scalar.

it points at: ax1.bar(pos, vals, width=width,label=cond,color=cm.accent(float(i)/n)).

i understand problem vals because should either scalar or have length 2, don't know how solve it. elements of vals float!

import numpy np import matplotlib.pyplot plt pylab import *  myarray=np.array([['series1', 'a',45],            ['series2', 'a',47],            ['series3', 'a',48],            ['series4','a',48],            ['series5', 'a',49],            ['series1','b',39],            ['series2','b',37],            ['series3','b',38],            ['series4','b',36],            ['series5','b',38]])  fig1=plt.figure() ax1=fig1.add_subplot(111) space=0.25  slots=np.unique(myarray[:,0]) categories=np.unique(myarray[:,1])  n=len(slots) width = (1 - space) / (len(slots))  i,cond in enumerate(slots[::-1]):     print "cond:", cond     vals = myarray[myarray[:,0] == cond][:,2]     pos = [j - (1 - space) / 2. + * width j in range(1,len(categories)+1)]     print(float(i)/n)     ax1.bar(pos, vals, width=width,label=cond,color=cm.accent(1-float(i+1)/n))  plt.show()