Annotate bars with values on Pandas bar plots

Question

I was looking for a way to annotate my bars in a Pandas bar plot with the rounded numerical values from my DataFrame.

>>> df=pd.DataFrame({'A':np.random.rand(2),'B':np.random.rand(2)},index=['value1','value2'] )         
>>> df
                 A         B
  value1  0.440922  0.911800
  value2  0.588242  0.797366

I would like to get something like this:

I tried with this code sample, but the annotations are all centered on the x ticks:

>>> ax = df.plot(kind='bar') 
>>> for idx, label in enumerate(list(df.index)): 
        for acc in df.columns:
            value = np.round(df.ix[idx][acc],decimals=2)
            ax.annotate(value,
                        (idx, value),
                         xytext=(0, 15), 
                         textcoords='offset points')

Solution 1

You get it directly from the axes' patches:

for p in ax.patches:
    ax.annotate(str(p.get_height()), (p.get_x() * 1.005, p.get_height() * 1.005))

You'll want to tweak the string formatting and the offsets to get things centered, maybe use the width from p.get_width(), but that should get you started. It may not work with stacked bar plots unless you track the offsets somewhere.

Solution 2

As of matplotlib 3.4.0:

A new Axes.bar_label helper method has been added for auto-labeling bar charts.

For single-group bar charts, supply ax.containers[0]:

df = pd.DataFrame({'A': np.random.rand(2)}, index=['value1', 'value2'])
ax = df.plot.barh()

ax.bar_label(ax.containers[0])

For multi-group bar charts, iterate ax.containers:

df = pd.DataFrame({'A': np.random.rand(2), 'B': np.random.rand(2)}, index=['value1', 'value2'])
ax = df.plot.bar()

for container in ax.containers:
    ax.bar_label(container)

See matplotlib's bar label demos for comprehensive examples using the optional styling params:

Axes.bar_label(self, container, labels=None, *, fmt='%g', label_type='edge', padding=0, **kwargs)

Solution 3

Solution which also handles the negative values with sample float formatting.

Still needs tweaking offsets.

df=pd.DataFrame({'A':np.random.rand(2)-1,'B':np.random.rand(2)},index=['val1','val2'] )
ax = df.plot(kind='bar', color=['r','b']) 
x_offset = -0.03
y_offset = 0.02
for p in ax.patches:
    b = p.get_bbox()
    val = "{:+.2f}".format(b.y1 + b.y0)        
    ax.annotate(val, ((b.x0 + b.x1)/2 + x_offset, b.y1 + y_offset))

Solution 4

The ax gives us the size of the box.

x_position=##define a value
y_position=##define a value
for patch in ax.patches:
    b= patch.get_bbox()
    y_value=b.y1-b.y0
    ax.annotate(y_value, "x_position" , "y_position"))
plt.show()

for more clarity::
Bbox(x0=3.75, y0=0.0, x1=4.25, y1=868.0)
Bbox(x0=4.75, y0=0.0, x1=5.25, y1=868.0)
Bbox(x0=5.75, y0=0.0, x1=6.25, y1=1092.0)
Bbox(x0=6.75, y0=0.0, x1=7.25, y1=756.0)
Bbox(x0=7.75, y0=0.0, x1=8.25, y1=756.0)
Bbox(x0=8.75, y0=0.0, x1=9.25, y1=588.0)
Bbox(x0=3.75, y0=868.0, x1=4.25, y1=3724.0)
Bbox(x0=4.75, y0=868.0, x1=5.25, y1=3528.0)
Bbox(x0=5.75, y0=1092.0, x1=6.25, y1=3948.0)
Bbox(x0=6.75, y0=756.0, x1=7.25, y1=2884.0)
Bbox(x0=7.75, y0=756.0, x1=8.25, y1=3024.0)
Bbox(x0=0.75, y0=4004.0, x1=1.25, y1=4396.0)
Bbox(x0=1.75, y0=3668.0, x1=2.25, y1=4060.0)
Bbox(x0=2.75, y0=3864.0, x1=3.25, y1=4060.0)

this is the output of patch.get_bbox() in my program.
we can extract the bounding box details from here and manipulate for our requirement

View more solutions

Comments

[leroygr]

Thanks @TomAugsPurger, the following worked: for p in ax.patches: ax.annotate(np.round(p.get_height(),decimals=2), (p.get_x()+p.get_width()/2., p.get_height()), ha='center', va='center', xytext=(0, 10), textcoords='offset points')

[leroygr]

Another question: how would you handle bars with negative values? With the code above, the labels are always with positive coordinates, even if the bar value is negative.

[TomAugspurger]

Hmm, you'd think there would be an easy way. I bet there is, but you could use np.sign(df) (which will be 1 or -1) and multiply that by the label location. Seems too complicated though.

[Nickpick]

Is there a similar solution for horizontal bars: kind=barh?

[Kamil Sindi]

I found this to work for barh: ax.annotate(str(p.get_width()), (p.get_x() + p.get_width(), p.get_y()), xytext=(5, 10), textcoords='offset points')

[DaveL17]

Thanks @capitalistpug. I found that adding horizontalalignment made yours better still. ax.annotate(str(int(p.get_width())), (p.get_x() + p.get_width(), p.get_y()), xytext=(-2, 4), textcoords='offset points', horizontalalignment='right') +1

[meduz]

Great answer and works very fine with pandas

[Lucas Brito]

And related to line graphs?