Pandas/Matplotlib annotate on one column and label on another

Question

I would like to annotote my plot by year inside a scatterplot. In addition I would also like to label (add legend) on a different column in a pandas dataframe, in this case the column: ds ['Label']. I have managed to annotate the scatter with the years but I'm stuck on how to label the data from a different column.

Here is my example code

ds

   Label  Year   factor1   factor2   factor3   factor4
0      A  2013  0.318451  0.038893 -0.145478  0.023298
1      B  2013  0.327400 -0.083985 -0.164712 -0.216095
2      C  2013  0.262333  0.251492  0.095186 -0.062729
3      D  2013  0.035074 -0.044357 -0.464473 -0.096461
4      E  2013  0.214464 -0.131810  0.065335 -0.339014
5      F  2013 -0.456510  0.111790  0.358160  0.327663
6      A  2012  0.345147 -0.010345 -0.139058 -0.033598
7      B  2012  0.318605 -0.096974 -0.168039  0.240126
8      C  2012  0.387761  0.145134  0.025229 -0.009165
9      D  2012 -0.007707 -0.033737 -0.401118  0.147932
10     E  2012  0.204582 -0.112144  0.007970  0.367639
11     F  2012 -0.439852  0.128267  0.355429 -0.375302


ds.columnsx=ds['factor2']
y=ds['factor1']
colors = {'A': 'b','B': 'purple', 'C': 'r','D' : 'grey','E' : 'green', 'F' : 'magenta'}
size= 2 *500

x=df['factor2']
y=df['factor1']

labels=df['Year']

fig=figure(1, figsize=(10,8))
ax1 = fig.add_subplot(111)

ax1.scatter(x, y, s=size, alpha=0.7, label=labels, color=[colors[i] for i in           ds['Label']])
for label, x, y in zip(labels, x, y):
plt.annotate(label, xy = (x, y),fontsize = 15)
grid(True)

ax1.spines['bottom'].set_color('orange')
ax1.spines['left'].set_color('green')

ax1.xaxis.label.set_color('orange')
ax1.yaxis.label.set_color('green')
ax1.tick_params(axis='x', colors='k')

plt.title('Something', fontsize = 15)

plt.xlabel('Something')
plt.ylabel('Something')
plt.tight_layout()
plt.show()

example

Community · Accepted Answer · 2017-05-23 10:31:45Z

2

Your question is a bit unclear but I think you want a legend that corresponds to the labels in ds['Label']. The way to do that is to call ax1.scatter once for each set of points, as in this question. For instance:

colors = {'A': 'b','B': 'purple', 'C': 'r','D' : 'grey','E' : 'green', 'F' : 'magenta'}
size= 1000

fig1, ax1 = plt.subplots(figsize=(10,8))

for t in ('A','B','C'):
    ax1.scatter(ds[ds['Label']==t]['factor2'], ds[ds['Label']==t]['factor1'], 
                color=colors[t], 
                label=t,
                s=size, 
                alpha=0.7)

for label, x, y in zip(ds['Year'], ds['factor2'], ds['factor1']):
    ax1.annotate(label, xy = (x + 0.008, y - 0.003),fontsize = 15)
ax1.legend(markerscale=0.2)

will give you a labelled legend. Is that what you're looking for?

edited May 23, 2017 at 10:31

CommunityBot

11 silver badge

answered Dec 2, 2013 at 16:15

jmz

4,3681 gold badge30 silver badges27 bronze badges

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

2 Comments

jonas Over a year ago

Yes, Thank you so much!!!. Also , do you know a way to keep the legend "bubbles" small while increasing the bubble size inside the scatter?

jmz Over a year ago

Use the markerscale argument to the ax1.legend call. I've updated the code in my answer to include it.

Collectives™ on Stack Overflow

Pandas/Matplotlib annotate on one column and label on another

1 Answer 1

2 Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Linked

Related