Data is Beautiful

A practical book on data visualisation that shows you how to create static and interactive visualisations that are engaging and beautiful.

Get the book

Top Olympic Medal Earning Countries

Dr. Shahin Rostami

2020-06-27

7 min read

Source

Contents Chat Share Follow Download Source

Preamble
Introduction

The Dataset

Data Wrangling
Chord Diagram
Chord Diagram with Names
Conclusion

Made with Plotapi

You can create beautiful, interactive, and engaging visualisations like this one with Plotapi in any programming language. Learn how to make beautiful visualisations with the book, Data is Beautiful.

Chord Diagram

Preamble¶

In [1]:

import numpy as np                   # for multi-dimensional containers 
import pandas as pd                  # for DataFrames
import itertools
from chord import Chord

Introduction¶

In previous sections, we visualised co-occurrences of Pokémon type. Whilst it was interesting to look at, the dataset only contained Pokémon from the first six geerations. In this section, we're going to use the TidyTuesday Animal Crossing villagers dataset to visualise the relationship between Species and .

The Dataset¶

The dataset documentation states that we can expect 13 variables per each of the 1017 Pokémon of the first eight generations.

Let's download the mirrored dataset and have a look for ourselves.

In [2]:

data_url = 'https://shahinrostami.com/datasets/athlete_events.csv'
raw_data = pd.read_csv(data_url)
raw_data.head()

Out[2]:

	ID	Name	Sex	Age	Height	Weight	Team	NOC	Games	Year	Season	City	Sport	Event	Medal
0	1	A Dijiang	M	24.0	180.0	80.0	China	CHN	1992 Summer	1992	Summer	Barcelona	Basketball	Basketball Men's Basketball	NaN
1	2	A Lamusi	M	23.0	170.0	60.0	China	CHN	2012 Summer	2012	Summer	London	Judo	Judo Men's Extra-Lightweight	NaN
2	3	Gunnar Nielsen Aaby	M	24.0	NaN	NaN	Denmark	DEN	1920 Summer	1920	Summer	Antwerpen	Football	Football Men's Football	NaN
3	4	Edgar Lindenau Aabye	M	34.0	NaN	NaN	Denmark/Sweden	DEN	1900 Summer	1900	Summer	Paris	Tug-Of-War	Tug-Of-War Men's Tug-Of-War	Gold
4	5	Christine Jacoba Aaftink	F	21.0	185.0	82.0	Netherlands	NED	1988 Winter	1988	Winter	Calgary	Speed Skating	Speed Skating Women's 500 metres	NaN

In [3]:

data = raw_data[raw_data.Medal.notna()]
data.head()

Out[3]:

	ID	Name	Sex	Age	Height	Weight	Team	NOC	Games	Year	Season	City	Sport	Event	Medal
3	4	Edgar Lindenau Aabye	M	34.0	NaN	NaN	Denmark/Sweden	DEN	1900 Summer	1900	Summer	Paris	Tug-Of-War	Tug-Of-War Men's Tug-Of-War	Gold
37	15	Arvo Ossian Aaltonen	M	30.0	NaN	NaN	Finland	FIN	1920 Summer	1920	Summer	Antwerpen	Swimming	Swimming Men's 200 metres Breaststroke	Bronze
38	15	Arvo Ossian Aaltonen	M	30.0	NaN	NaN	Finland	FIN	1920 Summer	1920	Summer	Antwerpen	Swimming	Swimming Men's 400 metres Breaststroke	Bronze
40	16	Juhamatti Tapio Aaltonen	M	28.0	184.0	85.0	Finland	FIN	2014 Winter	2014	Winter	Sochi	Ice Hockey	Ice Hockey Men's Ice Hockey	Bronze
41	17	Paavo Johannes Aaltonen	M	28.0	175.0	64.0	Finland	FIN	1948 Summer	1948	Summer	London	Gymnastics	Gymnastics Men's Individual All-Around	Bronze

capitalise the name, personality, and species of each villager.

It looks good so far, but let's confirm the 13 variables against 1017 samples from the documentation.

In [4]:

data.shape

Out[4]:

(39783, 15)

In [5]:

data = data[data['NOC'].isin(list(data['NOC'].value_counts()[:20].index))]

Perfect, that's exactly what we were expecting.

Data Wrangling¶

We need to do a bit of data wrangling before we can visualise our data. We can see from the columns names that the Pokémon types are split between the columns Type 1 and Type 2.

So let's select just these two columns and work with a list containing only them as we move forward.

In [6]:

species_personality = pd.DataFrame(data[['NOC', 'Medal']].values).dropna().astype(str)
species_personality

Out[6]:

	0	1
0	FIN	Bronze
1	FIN	Bronze
2	FIN	Bronze
3	FIN	Bronze
4	FIN	Gold
...	...	...
30152	URS	Gold
30153	URS	Silver
30154	URS	Bronze
30155	RUS	Bronze
30156	RUS	Silver

30157 rows × 2 columns

In [7]:

species_personality = species_personality.dropna()

In [ ]:

Now for the names of our types.

In [8]:

#left = np.unique(pd.DataFrame(species_personality)[0]).tolist()
left = list(data['Medal'].value_counts().index)[::-1]
#left.sort()
left = list(["Gold","Silver","Bronze"])

pd.DataFrame(left)

Out[8]:

	0
0	Gold
1	Silver
2	Bronze

In [9]:

#right = np.unique(pd.DataFrame(species_personality)[1]).tolist()
right = list(data['NOC'].value_counts().index)
#right.sort()
pd.DataFrame(right)

Out[9]:

	0
0	USA
1	URS
2	GER
3	GBR
4	FRA
5	ITA
6	SWE
7	CAN
8	AUS
9	RUS
10	HUN
11	NED
12	NOR
13	GDR
14	CHN
15	JPN
16	FIN
17	SUI
18	ROU
19	KOR

Which we can now use to create the matrix.

In [10]:

features= left+right
d = pd.DataFrame(0, index=features, columns=features)

Our chord diagram will need two inputs: the co-occurrence matrix, and a list of names to label the segments.

We can build a co-occurrence matrix with the following approach. We'll start by creating a list with every type pairing in its original and reversed form.

In [11]:

species_personality.values

Out[11]:

array([['FIN', 'Bronze'],
       ['FIN', 'Bronze'],
       ['FIN', 'Bronze'],
       ...,
       ['URS', 'Bronze'],
       ['RUS', 'Bronze'],
       ['RUS', 'Silver']], dtype=object)

In [12]:

for x in species_personality.values:
    d.at[x[0], x[1]] += 1
    d.at[x[1], x[0]] += 1

In [ ]:

Chord Diagram¶

Time to visualise the co-occurrence of types using a chord diagram. We are going to use a list of custom colours that represent the types.

In [13]:

colors =["#FFD700","#C0C0C0","#A57164",
'#e6194b', '#3cb44b', '#ffe119', '#4363d8', '#f58231', '#911eb4', '#46f0f0', '#f032e6', '#bcf60c', '#fabebe', '#008080', '#e6beff', '#9a6324', '#fffac8', '#800000', '#aaffc3', '#808000', '#ffd8b1', '#000075', '#808080', '#ffffff', '#000000'
         #'#e6194B', '#3cb44b', '#ffe119', '#4363d8', '#f58231', '#911eb4', '#42d4f4', '#f032e6', '#bfef45', '#fabed4', '#469990', '#dcbeff', '#9A6324', '#fffac8', '#800000', '#aaffc3', '#808000', '#ffd8b1', '#000075', '#a9a9a9', '#ffffff', '#000000'
        ]

In [14]:

names = left + right
len(names)

Out[14]:

Finally, we can put it all together.

In [29]:

Chord(d.values.tolist(), names,credit=True, colors=colors, curved_labels=False,
      margin=40, font_size_large=7,noun="medals", conjunction="won by", verb="",
        details_separator="", divide=True, divide_idx=len(left),divide_size=.2, reverse_gradients=True).show()

Chord(d.values.tolist(), names,credit=True, colors=colors, curved_labels=False,
      margin=40, font_size_large=7,noun="medals", conjunction="won by", verb="",
        details_separator="", divide=True, divide_idx=len(left),divide_size=.2, reverse_gradients=False).show()

Chord Diagram

Chord Diagram with Names¶

It would be nice to show a list of Pokémon names when hovering over co-occurring Pokémon types. To do this, we can make use of the optional details parameter.

In [ ]:

Next, we'll create an empty multi-dimensional array with the same shape as our matrix.

In [16]:

details = np.empty((len(names),len(names)),dtype=object)
details_thumbs = np.empty((len(names),len(names)),dtype=object)

Now we can populate the details array with lists of Pokémon names in the correct positions.

In [17]:

for count_x, item_x in enumerate(names):
    for count_y, item_y in enumerate(names):
        details_urls = data[
            (data['species'].isin([item_x, item_y])) &
            (data['personality'].isin([item_y, item_x]))]['url'].to_list()
        
        details_names = data[
            (data['species'].isin([item_x, item_y])) &
            (data['personality'].isin([item_y, item_x]))]['name'].to_list()
        
        urls_names = np.column_stack((details_urls, details_names))
        if(urls_names.size > 0):
            details[count_x][count_y] = details_names
            details_thumbs[count_x][count_y] = details_urls

        else:
            details[count_x][count_y] = []
            details_thumbs[count_x][count_y] = []

details=pd.DataFrame(details).values.tolist()
details_thumbs=pd.DataFrame(details_thumbs).values.tolist()

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
~/miniconda3/envs/analytics/lib/python3.9/site-packages/pandas/core/indexes/base.py in get_loc(self, key, method, tolerance)
   3079             try:
-> 3080                 return self._engine.get_loc(casted_key)
   3081             except KeyError as err:

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc()

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc()

pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item()

pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item()

KeyError: 'species'

The above exception was the direct cause of the following exception:

KeyError                                  Traceback (most recent call last)
<ipython-input-17-f54e3a0b2906> in <module>
      2     for count_y, item_y in enumerate(names):
      3         details_urls = data[
----> 4             (data['species'].isin([item_x, item_y])) &
      5             (data['personality'].isin([item_y, item_x]))]['url'].to_list()
      6 

~/miniconda3/envs/analytics/lib/python3.9/site-packages/pandas/core/frame.py in __getitem__(self, key)
   3022             if self.columns.nlevels > 1:
   3023                 return self._getitem_multilevel(key)
-> 3024             indexer = self.columns.get_loc(key)
   3025             if is_integer(indexer):
   3026                 indexer = [indexer]

~/miniconda3/envs/analytics/lib/python3.9/site-packages/pandas/core/indexes/base.py in get_loc(self, key, method, tolerance)
   3080                 return self._engine.get_loc(casted_key)
   3081             except KeyError as err:
-> 3082                 raise KeyError(key) from err
   3083 
   3084         if tolerance is not None:

KeyError: 'species'

In [ ]:

len(right)

Finally, we can put it all together but this time with the details matrix passed in.

In [ ]:

Chord(d.values.tolist(), names,credit=True, colors=colors, wrap_labels=False,
      margin=40, font_size_large=7,details=details,details_thumbs=details_thumbs,noun="villagers",
        details_separator="", divide=True, divide_idx=len(left),divide_size=.2, width=850).to_html()

In [ ]:

np.empty(shape=(6,1)).tolist()

Conclusion¶

In this section, we demonstrated how to conduct some data wrangling on a downloaded dataset to prepare it for a chord diagram. Our chord diagram is interactive, so you can use your mouse or touchscreen to investigate the co-occurrences!

Made with Plotapi

You can create beautiful, interactive, and engaging visualisations like this one with Plotapi in any programming language. Learn how to make beautiful visualisations with the book, Data is Beautiful.

Data is Beautiful

A practical book on data visualisation that shows you how to create static and interactive visualisations that are engaging and beautiful.

Get the book

	0
0	USA
1	URS
2	GER
3	GBR
4	FRA
5	ITA
6	SWE
7	CAN
8	AUS
9	RUS
10	HUN
11	NED
12	NOR
13	GDR
14	CHN
15	JPN
16	FIN
17	SUI
18	ROU
19	KOR

	0
0	USA
1	URS
2	GER
3	GBR
4	FRA
5	ITA
6	SWE
7	CAN
8	AUS
9	RUS
10	HUN
11	NED
12	NOR
13	GDR
14	CHN
15	JPN
16	FIN
17	SUI
18	ROU
19	KOR

	0
0	USA
1	URS
2	GER
3	GBR
4	FRA
5	ITA
6	SWE
7	CAN
8	AUS
9	RUS
10	HUN
11	NED
12	NOR
13	GDR
14	CHN
15	JPN
16	FIN
17	SUI
18	ROU
19	KOR