Create a graph (plot) on the data of the nested dictionary in matplotlib

Question

I can build a simple plot in a matplotlib according to the 'simple' dictionary:

import matplotlib.pyplot as plt
D = {u'Label1':26, u'Label2': 17, u'Label3':30}
plt.bar(range(len(D)), D.values(), align='center')
plt.xticks(range(len(D)), D.keys())

But, how do I create two graphics on the data of these dictionaries on one the plot I do not know?

NG1={'need1': [{'good1': 3, 'good2': 4}], 'need2': [{'good2': 2, 'good3': 2}]}
NG2={'need1': [{'good1': 13, 'good2': 23}], 'need2': [{'good2': 8, 'good3': 14}]}

Like the picture below

What is the desired result? What do you expect the graph to look like? — DavidG
– DavidG, Commented Oct 17, 2017 at 13:43
Hi! Tell me, this is possible to create or not, using matplotlib? — Yaroslav
– Yaroslav, Commented Oct 17, 2017 at 15:22

Maarten Fabré · Accepted Answer · 2017-10-17 15:32:30Z

5

using pandas you can do approximately what I think you want

NG1={'need1': {'good1': 3, 'good2': 4}, 'need2': {'good2': 2, 'good3': 2}}
NG2={'need1': {'good1': 13, 'good2': 23}, 'need2': {'good2': 8, 'good3': 14}}

(notice the lack of [])

combined_df = pd.concat({'ng1': pd.DataFrame(NG1), 'ng2': pd.DataFrame(NG2)}).unstack(0)

combined_df

   need1            need2
    ng1     ng2     ng1     ng2
good1   3.0     13.0    NaN     NaN
good2   4.0     23.0    2.0     8.0
good3   NaN     NaN     2.0     14.0

Depending on what exactly it is you want you can omit the unstack

combined_df.plot.bar()

delivers

edit update

I cannot create exactly what you need in this way, You would need to play around with different glyphs and figures and I don't have the skills or time to do this, but I can provide the data in the correct way

combined_df = pd.concat({'ng1': pd.DataFrame(NG1), 'ng2': pd.DataFrame(NG2)}).stack()
combined_df.index.names = ['ng', 'good', 'need']
combined_df = combined_df.unstack(['good'])
combined_df['sum'] = combined_df.sum(axis=1)

combined_df

    good    good1   good2   good3   sum
ng  need                
ng1     need1   3.0     4.0     NaN     7.0
need2   NaN     2.0     2.0     4.0
ng2     need1   13.0    23.0    NaN     36.0
need2   NaN     8.0     14.0    22.0

combined_df.plot.bar()

edited Oct 17, 2017 at 15:32

answered Oct 17, 2017 at 13:53

Maarten Fabré

7,0781 gold badge19 silver badges37 bronze badges

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

2 Comments

Yaroslav Over a year ago

Thank you for the answer! I add the question with the figure below. Tell me this is possible to create or not?

Yaroslav Over a year ago

And how to make two drawings : ng1 (need1, need2) and ng2 (need1, need2) on the one picture. Is this possible?

ImportanceOfBeingErnest · Accepted Answer · 2017-10-17 19:33:28Z

3

Adding to @MaartenFabré 's solution, you can get the sum as another bar plot in the background by plotting the sum as an individual bar plot and make it unfilled,

combined_df['sum'].plot.bar(zorder=0, fill=False)

Complete solution:

import matplotlib.pyplot as plt
import pandas as pd

NG1={'need1': {'good1': 3, 'good2': 4}, 'need2': {'good2': 2, 'good3': 2}}
NG2={'need1': {'good1': 13, 'good2': 23}, 'need2': {'good2': 8, 'good3': 14}}

combined_df = pd.concat({'ng1': pd.DataFrame(NG1), 'ng2': pd.DataFrame(NG2)}).stack()
combined_df.index.names = ['ng', 'good', 'need']
combined_df = combined_df.unstack(['good'])

combined_df.plot.bar()

combined_df['sum'] = combined_df.sum(axis=1)
combined_df['sum'].plot.bar(zorder=0, fill=False)


plt.show()

answered Oct 17, 2017 at 19:33

ImportanceOfBeingErnest

342k61 gold badges737 silver badges771 bronze badges

1 Comment

Yaroslav Over a year ago

And how to make two drawings : ng1 (need1, need2) and ng2 (need1, need2) on the one picture. Is this possible?

Collectives™ on Stack Overflow

Create a graph (plot) on the data of the nested dictionary in matplotlib

2 Answers 2

edit update

2 Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

edit update

2 Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related