Olympia 2016

Creating an interactive visualization for exploring the results of the 2016 Rio Olympics.

Importing libraries and packages

# Warnings
import warnings

# Mathematical operations and data manipulation
import pandas as pd
import random

# Visualisation
from bokeh.plotting import figure, show, ColumnDataSource

# https://ipywidgets.readthedocs.io/en/latest/examples/Using%20Interact.html
from ipywidgets import interact
import ipywidgets as widgets

# Output
from bokeh.io import output_notebook

output_notebook()
warnings.filterwarnings("ignore")
%matplotlib inline

Loading BokehJS ...

Set paths

# Path to datasets directory
data_path = "./datasets"
# Path to assets directory (for saving results to)
assets_path = "./assets"

Loading dataset

dataset = pd.read_csv(f"{data_path}/olympia2016_athletes.csv")

Exploring dataset

# Shape of the dataset
print("Shape of the dataset: ", dataset.shape)
# Head
dataset.head()

Shape of the dataset:  (11538, 11)

	id	name	nationality	sex	dob	height	weight	sport	gold	silver	bronze
0	736041664	A Jesus Garcia	ESP	male	10/17/69	1.72	64.0	athletics	0	0	0
1	532037425	A Lam Shin	KOR	female	9/23/86	1.68	56.0	fencing	0	0	0
2	435962603	Aaron Brown	CAN	male	5/27/92	1.98	79.0	athletics	0	0	1
3	521041435	Aaron Cook	MDA	male	1/2/91	1.83	80.0	taekwondo	0	0	0
4	33922579	Aaron Gate	NZL	male	11/26/90	1.81	71.0	cycling	0	0	0

Preprocessing

# extract countries and group olympians by country
# and the number of medals per country
countries = dataset["nationality"].unique()
athletes_per_country = dataset.groupby("nationality").size()
medals_per_country = dataset.groupby("nationality")[
    "gold", "silver", "bronze"
].sum()

Visualisation

# creating the scatter plot
def get_plot(max_athletes, max_medals):
    filtered_countries = []

    for country in countries:
        if (
            athletes_per_country[country] <= max_athletes
            and medals_per_country.loc[country].sum() <= max_medals
        ):
            filtered_countries.append(country)

    data_source = get_datasource(filtered_countries)
    TOOLTIPS = [
        ("Country", "@countries"),
        ("Num of Athletes", "@y"),
        ("Gold", "@gold"),
        ("Silver", "@silver"),
        ("Bronze", "@bronze"),
    ]

    plot = figure(
        title="Rio Olympics 2016 - Medal comparison",
        x_axis_label="Number of Medals",
        y_axis_label="Num of Athletes",
        plot_width=800,
        plot_height=500,
        tooltips=TOOLTIPS,
    )

    plot.circle(
        "x", "y", source=data_source, size=20, color="color", alpha=0.5
    )

    return plot

# get a 6 digit random hex color to differentiate the countries better
def get_random_color():
    return "#%06x" % random.randint(0, 0xFFFFFF)

# build the datasource
def get_datasource(filtered_countries):
    return ColumnDataSource(
        data=dict(
            color=[get_random_color() for _ in filtered_countries],
            countries=filtered_countries,
            gold=[
                medals_per_country.loc[country]["gold"]
                for country in filtered_countries
            ],
            silver=[
                medals_per_country.loc[country]["silver"]
                for country in filtered_countries
            ],
            bronze=[
                medals_per_country.loc[country]["bronze"]
                for country in filtered_countries
            ],
            x=[
                medals_per_country.loc[country].sum()
                for country in filtered_countries
            ],
            y=[
                athletes_per_country.loc[country].sum()
                for country in filtered_countries
            ],
        )
    )

# getting the max amount of medals and athletes of all countries
max_medals = medals_per_country.sum(axis=1).max()
max_athletes = athletes_per_country.max()

# setting up the interaction elements
max_athletes_slider = widgets.IntSlider(
    value=max_athletes,
    min=0,
    max=max_athletes,
    step=1,
    description="Max. Athletes:",
    continuous_update=False,
    orientation="vertical",
    layout={"width": "100px"},
)

max_medals_slider = widgets.IntSlider(
    value=max_medals,
    min=0,
    max=max_medals,
    step=1,
    description="Max. Medals:",
    continuous_update=False,
    orientation="horizontal",
)

# creating the interact method
@interact(max_athletes=max_athletes_slider, max_medals=max_medals_slider)
def get_olympia_stats(max_athletes, max_medals):
    show(get_plot(max_athletes, max_medals))