using a list of colors when plotting multiple series with matplotlib

Question

I would like to add multiple columns of a pandas DataFrame to a matplotlib axis and define the colors using a list of color names. When passing a list to the color argument I get a value error: Invalid RGBA argument. The following MWE reproduces this error:

import pandas as pd
from matplotlib import pyplot as plt
import matplotlib.patches as mpatches

df = pd.DataFrame({'0':[0,1,0],'a':[1,2,3],'b':[2,4,6],'c':[5,3,1]})
colors = ['r','g','b']
fig, ax = plt.subplots()
ax.bar(df.index.values,df['0'].values, color = 'y')
ax2 = ax.twinx()
h = ax2.plot(df.index.values, df[['a','b','c']].values, color = colors)
handles = [mpatches.Patch(color='y')]
handles = handles + h
labels = df.columns()
lgd=ax.legend(handles,labels,loc='center left', bbox_to_anchor=(1.1, 0.5), ncol=1, fancybox=True, shadow=True, fontsize=ls)
plt.savefig('test.png', bbox_extra_artists=(lgd,tbx), bbox_inches='tight')

Answer 1

The matplotlib plot's color argument only accepts a single color. Options:

• An easy option is to use a color cycler instead

  ax2.set_prop_cycle('color',colors )
  h = ax2.plot(df.index.values, df[['a','b','c']].values)
  

• You could also loop over the lines after plotting,

  h = ax2.plot(df.index.values, df[['a','b','c']].values)
  for line, color in zip(h,colors):
      line.set_color(color)
  

• Finally consider using the pandas plotting wrapper,

  df[['a','b','c']].plot(ax = ax2, color=colors)
  

All options result in the same plot.