Matplotlib错误栏无法读取熊猫数据框

Question

I have two data frames in pandas format that I am trying to plot as values and error bars. But the python interface complains about some error I cannot understand. I have tested a colleague's almost the same code, and it appears that the fact that I run python 3.5 while he utilizes 2.7, is the source of the error. Therefore, I did test his code on my computer (python 3.5) and I am getting the same error message.

Bellow is a subset of my troubling code:

"Using pandas library to combine the three white spruce data sets"
trees = [white_spruce_1,white_spruce_2,white_spruce_3]
ntrees = pd.concat(trees) # Concatenate list into a series
spruce_stat = ntrees.groupby("Wvl") #Converted the series into a panda object

mean_spruce = spruce_stat.mean()
std_spruce = spruce_stat.std()
#mean_spruce.head()
mean_spruce['wvl']=mean_spruce.index
mean_spruce.head()

     Chan.# Rad. (Ref.) Rad. (Target)   Tgt./Ref. %
Wvl             
350     0        0       0.000014        0.686176
351     0        0       0.000015        0.707577

std_spruce.head()

       Chan.#   Rad. (Ref.) Rad. (Target)   Tgt./Ref. %
Wvl             
350       0       0          0.000014        0.686176
351       0       0          0.000015        0.707577

plt.errorbar(mean_spruce['wvl'],mean_spruce['Tgt./Ref. %'], xerr = None, yerr = std_spruce['Rad. (Ref.)'])

Bellow is the error message I receive:

KeyError                Traceback (most recent call last)
<ipython-input-52-13352d94b09c> in <module>()
      2 #plt.errorbar(mean_spruce['wvl'],mean_spruce['Tgt./Ref. %'], xerr = None,yerr=std_spruce['Tgt./Ref. %'],c='k',ecolor='r', elinewidth=0.5, errorevery=5)
      3 #plt.errorbar( x, y, xerr = None , yerr = sd_white_spruce['Tgt./Ref. %'],c = 'green', ecolor = 'red', capsize = 0,elinewidth = 0.5, errorevery = 5 )
----> 4 plt.errorbar(mean_spruce['wvl'],mean_spruce['Tgt./Ref. %'], xerr = None, yerr = std_spruce['Rad. (Ref.)'])# ,c = 'green', ecolor = 'red', capsize = 0,elinewidth = 0.5, errorevery = 5)
      5 

C:\Users\mike\Anaconda3\lib\site-packages\matplotlib\pyplot.py in errorbar(x, y, yerr, xerr, fmt, ecolor, elinewidth, capsize, barsabove, lolims, uplims, xlolims, xuplims, errorevery, capthick, hold, data, **kwargs)
   2828                           xlolims=xlolims, xuplims=xuplims,
   2829                           errorevery=errorevery, capthick=capthick, data=data,
-> 2830                           **kwargs)
   2831     finally:
   2832         ax.hold(washold)

C:\Users\mike\Anaconda3\lib\site-packages\matplotlib\__init__.py in inner(ax, *args, **kwargs)
   1809                     warnings.warn(msg % (label_namer, func.__name__),
   1810                                   RuntimeWarning, stacklevel=2)
-> 1811             return func(ax, *args, **kwargs)
   1812         pre_doc = inner.__doc__
   1813         if pre_doc is None:

C:\Users\mike\Anaconda3\lib\site-packages\matplotlib\axes\_axes.py in errorbar(self, x, y, yerr, xerr, fmt, ecolor, elinewidth, capsize, barsabove, lolims, uplims, xlolims, xuplims, errorevery, capthick, **kwargs)
   2961                 # Check for scalar or symmetric, as in xerr.
   2962                 if len(yerr) > 1 and not ((len(yerr) == len(y) and not (
-> 2963                         iterable(yerr[0]) and len(yerr[0]) > 1))):
   2964                     raise ValueError("yerr must be a scalar, the same "
   2965                                      "dimensions as y, or 2xN.")

C:\Users\mike\Anaconda3\lib\site-packages\pandas\core\series.py in __getitem__(self, key)
    555     def __getitem__(self, key):
    556         try:
--> 557             result = self.index.get_value(self, key)
    558 
    559             if not np.isscalar(result):

C:\Users\mike\Anaconda3\lib\site-packages\pandas\core\index.py in get_value(self, series, key)
   3882 
   3883         k = _values_from_object(key)
-> 3884         loc = self.get_loc(k)
   3885         new_values = _values_from_object(series)[loc]
   3886 

C:\Users\mike\Anaconda3\lib\site-packages\pandas\core\index.py in get_loc(self, key, method, tolerance)
   3940             pass
   3941         return super(Float64Index, self).get_loc(key, method=method,
-> 3942                                                  tolerance=tolerance)
   3943 
   3944     @property

C:\Users\mike\Anaconda3\lib\site-packages\pandas\core\index.py in get_loc(self, key, method, tolerance)
   1757                                  'backfill or nearest lookups')
   1758             key = _values_from_object(key)
-> 1759             return self._engine.get_loc(key)
   1760 
   1761         indexer = self.get_indexer([key], method=method,

pandas\index.pyx in pandas.index.IndexEngine.get_loc (pandas\index.c:3979)()

pandas\index.pyx in pandas.index.IndexEngine.get_loc (pandas\index.c:3843)()

pandas\hashtable.pyx in pandas.hashtable.Float64HashTable.get_item (pandas\hashtable.c:9556)()

pandas\hashtable.pyx in pandas.hashtable.Float64HashTable.get_item (pandas\hashtable.c:9494)()

KeyError: 0.0

Thanks for the help

Answer 1

The problem is the discrepancy between the Pandas indexing and the indexing of Matplotlib internal functions. One way to resolve it -albeit not being elegant- is to create a dummy dataframe just for the plotting purpose. In your case:

mean_spruce_dummy = mean_spruce
mean_spruce_dummy.columns = np.arange(0, len(mean_spruce))

In principle, this discrepancy is solved in the newer version of Pandas.

Answer 2

I'm seeing a similar error in python 2.7. My solution is to access the underlying data directly. This should work for you

x = mean_spruce['wvl'].values
y = mean_spruce['Tgt./Ref. %'].values
yerr = std_spruce['Rad. (Ref.)'].values

plt.errorbar(x, y yerr=yerr)