Pandas subplot using two series

Question

I have two series' which contains the same data, but they contain a different number of occurrences of this data. I want to compare these two series' by making a bar chart, where the two are compared. Below is what I've done so far.

import matplotlib.patches as mpatches

fig = plt.figure()

ax = fig.add_subplot(111)

width = 0.3

tree_amount15.plot(kind='bar', color='red', ax=ax, width=width, position=1, label='NYC')
queens_tree_types.plot(kind='bar', color='blue', ax=ax, width=width, position=0, label='Queens')
plt.legend(bbox_to_anchor=(0., 1.02, 1., .102), loc=3,
       ncol=2, mode="expand", borderaxespad=0.)

ax.set_ylabel('Total trees')
ax.set_xlabel('Tree names')

plt.show()

Which gives me the following chart:

The problem I have is that, even though all the 'Tree names' are the same in each series, the 'Total trees' is of course different, so for example, #5 (Callery pear) is only #5 in 'tree_amount15', where it's #3 in 'queens_tree_types' and so on. How can I order the series such that it's the value that corresponds to the right label shown on the chart? Because right now, it's the labels from the series that gets added first, which is shown, which makes the values of the second series be misleading.

Any hints?

Here's how the two series look, when I do a value_counts() them.

tree_amount15:

London planetree     87014
honeylocust          64264
Callery pear         58931
pin oak              53185
Norway maple         34189
littleleaf linden    29742
cherry               29279
Japanese zelkova     29258
ginkgo               21024
Sophora              19338
red maple            17246
green ash            16251
American linden      13530
silver maple         12277
sweetgum             10657
northern red oak      8400
silver linden         7995
American elm          7975
maple                 7080
purple-leaf plum      6879

queens_tree_types:

London planetree     31111
pin oak              22610
honeylocust          20290
Norway maple         19407
Callery pear         16547
cherry               13497
littleleaf linden    11902
Japanese zelkova      8987
green ash             7389
silver maple          6116
ginkgo                5971
Sophora               5386
red maple             4935
American linden       4769
silver linden         4146
purple-leaf plum      3035
maple                 2992
northern red oak      2697
sweetgum              2489
American elm          1709

Are you saying that the blue bar corresponding to Callery pear is incorrect, and should be the blue bar from the honeylocust column? — James
– James, Commented Apr 14, 2017 at 13:23
The labels correspond to the blue bar correctly, though the red bar does not. It's as if it's just plotted the two series in whatever order they're in, and then added the labels for the series added last. This means the red bar doesn't correspond with the labels. — Khaine775
– Khaine775, Commented Apr 14, 2017 at 13:25
Can you please post an example of your data from tree_amount15 and queens_tree_types? Are these pandas series or dataframe objects? — James
– James, Commented Apr 14, 2017 at 13:28

James · Accepted Answer · 2017-04-14 14:04:59Z

2

You can create a data frame from your two series that uses the tree name index. By default pandas will sort the index alphabetically, so we tell it to sort using the values of NYC. With both series as columns, we can use a single call to the plot method to put them on the same graph.

df = pd.concat([tree_amount15, queens_tree_types], axis=1).rename_axis(
          {0:'NYC', 1:'Queens'}, axis='columns') # sets the column names

df.sort_values('NYC', ascending=False)           # sort the df using NYC values

df.plot.bar(color=['red','blue'])

answered Apr 14, 2017 at 14:04

James

37k4 gold badges54 silver badges79 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Khaine775 Over a year ago

Neat, thanks. For some reason, the renaming didn't work, so I removed that and did df.columns = ['NYC', 'Queens'] instead!

Collectives™ on Stack Overflow

Pandas subplot using two series

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related