Bar graph plot with values on top python

Question

I have a table in my pandas df.

    Total_orders    frequency
      0.0           18679137
      1.0           360235
      2.0           68214
      3.0           20512
      4.0           7211
      ...           ...
      50.0          12

I want to plot a bar graph total orders vs frequency, with the values of frequency displayed on the top of each bars.

I am running these three codes.

Code1:

plt.figure(figsize=(12,6))
df2 = df.groupby('Total_orders')['frequency'].plot(kind='bar')
plt.xlabel('Total_orders')
plt.ylabel('frequency')

for rect in df2.patches:
    height = rect.get_height()
    df2.text(rect.get_x() + rect.get_width()/2., 1.05*height+100,
'%d' % int(height),ha='center', va='bottom', rotation=90)

Code2:(for loop)

for ii,rect in enumerate(df2.patches):
    height = rect.get_height()
    df2.text(rect.get_x() + rect.get_width()/2., 1.14*height+100,'%d' % int(height),ha='center', va='bottom', rotation=90)

Code3

for p in df2.patches:
df2.annotate(str(p.get_height()), (p.get_x() * 1.005, p.get_height() *1.05),rotation=90)

but when I am running the code it's showing me error, that

AttributeError: 'Series' object has no attribute 'patches'

Any idea why this is happening, and how to remove it?

Answer 1

Try this:

def autolabel(rects, height_factor=1.05):
    # attach some text labels
    for rect in rects:
        height = rect.get_height()
        ax.text(rect.get_x() + rect.get_width()/2., height_factor*height,
                '%d' % int(height),
                ha='center', va='bottom')


In [48]: df
Out[48]:
   Total_orders  frequency
0           0.0   18679137
1           1.0     360235
2           2.0      68214
3           3.0      20512
4           4.0       7211
5          50.0         12

In [49]: import matplotlib

In [50]: matplotlib.style.use('ggplot')

In [51]: ax = df.plot.bar(x='Total_orders', y='frequency', rot=0, width=0.85, alpha=0.6, figsize=(14,12))

In [52]: autolabel(ax.patches, 1.02)