Change subplot color in DataFrame?

Question

I would like to change color of individual subplot:
1. Specifying desired color of plots by hand
2. Using random colors

Basic code (taken from 1)

 df = DataFrame(np.random.randn(1000, 4), index=ts.index, columns=list('ABCD'))
 df = df.cumsum()

 df.plot(subplots=True)
 plt.legend(loc='best') 
 plt.show()

I tried this:

 colors = ['r','g','b','r']                   #first option
 colors = list(['r','g','b','r'])             #second option
 colors = plt.cm.Paired(np.linspace(0,1,4))   #third option

 df.plot(subplots=True, color=colors)

But all of them doesn' work. I found 2 but I'm not sure how to change this:

 plots=df.plot(subplots=True)
 for color in plots:
   ??????

Answer 1

You can easlity do this by providing the style parameter with a list of color abbrevations:

from pandas import Series, DataFrame, date_range
import matplotlib.pyplot as plt
import numpy as np

ts = Series(np.random.randn(1000), index=date_range('1/1/2000', periods=1000))
ts = ts.cumsum()

df = DataFrame(np.random.randn(1000, 4), index=ts.index, columns=list('ABCD'))
df = df.cumsum()

ax = df.plot(subplots=True, style=['r','g','b','r'], sharex=True)
plt.legend(loc='best')
plt.tight_layout()
plt.show()

---

Random colors from "standard" colors

If you only want to use the "standard" colors (blue, green, red, cyan, magenta, yellow, black, white), you can define an array containing the color abbrevations and pass a random sequence of these as argument to the style parameter:

colors = np.array(list('bgrcmykw'))

...

ax = df.plot(subplots=True,
             style=colors[np.random.randint(0, len(colors), df.shape[1])],
             sharex=True) 

Answer 2

You'll have to loop over each axis by itself and do this manually as far as I can tell. This should probably be a feature.

df = DataFrame(np.random.randn(1000, 4), columns=list('ABCD'))
df = df.cumsum()
colors = 'r', 'g', 'b', 'k'
fig, axs = subplots(df.shape[1], 1,sharex=True, sharey=True)
for ax, color, (colname, col) in zip(axs.flat,colors,df.iteritems()):
    ax.plot(col.index,col,label=colname,c=color)
    ax.legend()
    ax.axis('tight')
fig.tight_layout()

Or if you have dates in your index do:

import pandas.util.testing as tm

df = tm.makeTimeDataFrame()
df = df.cumsum()
colors = 'r', 'g', 'b', 'k'
fig, axs = subplots(df.shape[1], 1,sharex=True, sharey=True)
for ax, color, (colname, col) in zip(axs.flat, colors, df.iteritems()):
    ax.plot(col.index, col, label=colname,c=color)
    ax.legend()
    ax.axis('tight')
fig.tight_layout()
fig.autofmt_xdate()

to get