How to plot a heatmap from pandas DataFrame

Question

Here is my dataframe:

              jan          f            m        a        m         j  \\
2000    -7.894737  22.387006    22.077922  14.5455  15.8038           


        

Answer 1

That is straightforward using seaborn; I demonstrate how to do it using random data, so all you have to do is to replace data in the example below by your actual dataframe.

My dataframe looks like this:

             A         B         C         D         E
2000  0.722553  0.948447  0.598707  0.656252  0.618292
2001  0.920532  0.054941  0.909858  0.721002  0.222167
2002  0.048496  0.963871  0.689730  0.697573  0.349308
2003  0.692897  0.272768  0.581736  0.150674  0.861672
2004  0.889694  0.658286  0.879855  0.739821  0.010971
2005  0.937347  0.132955  0.704528  0.443084  0.552123
2006  0.869499  0.750177  0.675160  0.873720  0.270204
2007  0.156933  0.186630  0.371993  0.153790  0.397232
2008  0.384696  0.585156  0.746883  0.185457  0.095387
2009  0.667236  0.340058  0.446081  0.863402  0.227776
2010  0.817394  0.343427  0.804157  0.245394  0.850774

The output then looks as follows (please note that the index is at the x-axis and the column names at the y-axis as requested):

Here is the entire code with some inline comments:

import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd

# create some random data; replace that by your actual dataset
data = pd.DataFrame(np.random.rand(11, 5), columns=['A', 'B', 'C', 'D', 'E'], index = range(2000, 2011, 1))

# plot heatmap
ax = sns.heatmap(data.T)

# turn the axis label
for item in ax.get_yticklabels():
    item.set_rotation(0)

for item in ax.get_xticklabels():
    item.set_rotation(90)

# save figure
plt.savefig('seabornPandas.png', dpi=100)
plt.show()