app.py

import json
from collections import Counter
import numpy as np
import pandas as pd
import altair as alt
import operator
import panel as pn
import warnings
warnings.filterwarnings('ignore')

pn.extension('vega')

# loading the events data
events={}
nations = ['Italy','England','Germany','France','Spain','European_Championship','World_Cup']
for nation in nations:
    with open('events/events_%s.json' %nation) as json_data:
        events[nation] = json.load(json_data)
        
# loading the match data
matches={}
nations = ['Italy','England','Germany','France','Spain','European_Championship','World_Cup']
for nation in nations:
    with open('matches/matches_%s.json' %nation) as json_data:
        matches[nation] = json.load(json_data)

# loading the players data
players={}
with open('players.json') as json_data:
    players = json.load(json_data)

# loading the competitions data
competitions={}
with open('competitions.json') as json_data:
    competitions = json.load(json_data)

# drawing the first chart
ev_all_nations = []
for nation in nations:
    for i in range(len(events[nation])):
        ev_all_nations.append(events[nation][i]['eventName'])
count = Counter(ev_all_nations)
total = len(ev_all_nations)

counter = {event: int((count / total) * 100) for event, count in count.items()}
sorted_counter = sorted(counter.items(), key=operator.itemgetter(1), reverse=False)

data = pd.DataFrame(sorted_counter, columns=['Event', 'Percentage'])

brush = alt.selection_interval(encodings=['y'])

max_value = data['Percentage'].max()
tick_values = list(range(0, int(max_value) + 10, 10))

bars = alt.Chart(data).mark_bar().encode(
    y=alt.Y('Event:N', title=None, sort='-x'),
    x=alt.X('Percentage:Q', title='events(%)', axis=alt.Axis(values=tick_values)),
    color=alt.condition(brush, alt.Color('Event:N', legend=None), alt.value('lightgray'))
).add_selection(
    brush
)

average_rule = alt.Chart(data).mark_rule(color='firebrick', strokeWidth=2).encode(
    x='mean(Percentage):Q'
).transform_filter(
    brush
)

average_text = alt.Chart(data).mark_text(
    dx=5, dy=-5, color='firebrick', align='left', fontWeight='bold'
).encode(
    x=alt.X('mean(Percentage):Q', aggregate='mean'),
    text=alt.Text('mean(Percentage):Q', aggregate='mean', format='.1f')
).transform_filter(
    brush
)

chart1 = alt.layer(bars, average_rule, average_text).properties(
    width=600,
    height=500,
    title='Events Distribution'
)

# Display the combined chart
chart1

# drawing the second chart
match_ev_count = {}
for nation in nations:
    for ev in events[nation]:
        if ev['matchId'] not in match_ev_count:
            match_ev_count[ev['matchId']] = 1
        else:
            match_ev_count[ev['matchId']] += 1

data = pd.DataFrame({
    'Event Count': list(match_ev_count.values())
})

mean_val = int(np.mean(data['Event Count']))
std_val = int(np.std(data['Event Count']))

hist = alt.Chart(data).mark_bar().encode(
    alt.X('Event Count:Q', bin=alt.Bin(maxbins=20), title='events (n)', 
          axis=alt.Axis(values=np.arange(0, max(data['Event Count']) + 100, 100))),
    alt.Y('count()', title='frequency (n)')
).properties(
    width=600,
    height=400
)

text = alt.Chart(pd.DataFrame({'x': [mean_val + std_val], 'y': [1], 'text': [f'μ = {mean_val} \n σ = {std_val}']})).mark_text(
    align='left',
    baseline='top',
    fontSize=20,
    dx=-120,
    dy=-300
).encode(
    x='x:Q',
    y='y:Q',
    text='text:N'
)

chart2 = hist + text

# Display the combined chart
chart2.display()

# drawing the third chart
combined_chart1 = alt.hconcat(
    chart1,
    chart2,
    spacing=10
).resolve_scale(
    color='independent'
)
combined_chart1

# drawing the fourth chart
match_id = 2576335
a_match = []
for nation in nations:
    for ev in events[nation]:
        if ev['matchId'] == match_id:
            a_match.append(ev)

for nation in nations:
    for match in matches[nation]:
        if match['wyId'] == match_id:
            match_f = match

df_a_match = pd.DataFrame(a_match)

background_data = pd.DataFrame({
    'x': [0],
    'y': [0],
    'x2': [100],
    'y2': [100]
})

# Create the background
background = alt.Chart(background_data).mark_rect(
    color='#195905'
).encode(
    x='x:Q',
    y='y:Q',
    x2='x2:Q',
    y2='y2:Q'
)


#Define the center circle
center_circle = alt.Chart(pd.DataFrame({'x': [50], 'y': [50]})).mark_point(
    size=12000,
    color='white',
    strokeWidth=3
).encode(
    x='x:Q',
    y='y:Q'
)


# Create the border lines
border_lines_data = pd.DataFrame({
    'x': [1, 1, 99.5, 99.5, 1],
    'y': [1, 99.5, 99.5, 1, 1],
    'x2': [1, 99.5, 99.5, 1, 1],
    'y2': [99.5, 99.5, 1, 1, 1]
})


border_lines = alt.Chart(border_lines_data).mark_line(
    color='white',
    strokeWidth=3
).encode(
    x=alt.X('x:Q', scale=alt.Scale(domain=[1, 99.5])),
    y=alt.Y('y:Q', scale=alt.Scale(domain=[1, 99.5])),
    x2='x2:Q',
    y2='y2:Q'
)

midline_data = pd.DataFrame({
    'x': [50, 50,],
    'y': [1, 99, ]
})

# Create the line using `mark_line`
midline = alt.Chart(midline_data).mark_line(
    color='white',
    strokeWidth=3
).encode(
    x='x:Q',
    y='y:Q'
)
lines_data = pd.DataFrame({
    'x': [1, 17.5, 17.5, 1, 82.5,  82.5, 99,1,6.5,6.5,1, 99,93.5,93.5],
    'y': [21.3, 21.3, 77.7, 77.7, 21.3, 77.7, 77.7,37.5,37.5,62.5,62.5,37.5,37.5,62.5],
    'x2': [17.5, 17.5, 1, 17.5, 99,  82.5, 82.5, 6.5,6.5,1,6.5,93.5,93.5,99],
    'y2': [21.3, 77.7, 77.7, 77.7, 21.3, 21.3,77.7,37.5,62.5,62.5,62.5,37.5,62.5,62.5]
})

lines = alt.Chart(lines_data).mark_line(
    color='white',
    strokeWidth=3
).encode(
    x='x:Q',
    y='y:Q',
    x2='x2:Q',
    y2='y2:Q'
)

dot_positions = pd.DataFrame({
    'x': [12, 87],
    'y': [50, 50]
})

# Create the white dots
white_dots = alt.Chart(dot_positions).mark_point(
    size=100,
    color='white',
    filled=True
).encode(
    x='x:Q',
    y='y:Q'
)



theta = np.linspace(0, np.pi, 100)
semicircle_x = 12 + 9.5 * np.cos(theta) 
semicircle_y = 50 + 9.5 * np.sin(theta)

semicircle_data = pd.DataFrame({
    'x': semicircle_x,
    'y': semicircle_y
})

semicircle_data = semicircle_data[semicircle_data['x'] >= 17.5]

arc1 = alt.Chart(semicircle_data).mark_line(
    color='white',
    strokeWidth=3
).encode(
    x=alt.X('x', scale=alt.Scale(domain=[0, 100])),
    y=alt.Y('y', scale=alt.Scale(domain=[0, 100]))
)



theta = np.linspace(0, np.pi, 100)
semicircle_x2 = 12 + 9.5 * np.cos(theta)
semicircle_y2 = 50 - 9.5 * np.sin(theta)

semicircle_data2 = pd.DataFrame({
    'x': semicircle_x2,
    'y': semicircle_y2
})


semicircle_data2 = semicircle_data2[semicircle_data2['x'] >= 17.5]


arc2 = alt.Chart(semicircle_data2).mark_line(
    color='white',
    strokeWidth=3
).encode(
    x=alt.X('x', scale=alt.Scale(domain=[0, 100])),
    y=alt.Y('y', scale=alt.Scale(domain=[0, 100]))
)


theta = np.linspace(0, np.pi, 100)
semicircle_x3 = 87 - 9.5 * np.cos(theta)
semicircle_y3 = 50 + 9.5 * np.sin(theta)


semicircle_data3 = pd.DataFrame({
    'x': semicircle_x3,
    'y': semicircle_y3
})


semicircle_data3 = semicircle_data3[semicircle_data3['x'] <= 82.5]


arc3 = alt.Chart(semicircle_data3).mark_line(
    color='white',
    strokeWidth=3
).encode(
    x=alt.X('x', scale=alt.Scale(domain=[0, 100])),
    y=alt.Y('y', scale=alt.Scale(domain=[0, 100]))
)



theta = np.linspace(0, np.pi, 100)
semicircle_x4 = 87 - 9.5 * np.cos(theta)
semicircle_y4 = 50 - 9.5 * np.sin(theta)

semicircle_data4 = pd.DataFrame({
    'x': semicircle_x4,
    'y': semicircle_y4
})

semicircle_data4 = semicircle_data4[semicircle_data4['x'] <= 82.5]

arc4 = alt.Chart(semicircle_data4).mark_line(
    color='white',
    strokeWidth=3
).encode(
    x=alt.X('x', scale=alt.Scale(domain=[0, 100]), title=None ),
    y=alt.Y('y', scale=alt.Scale(domain=[0, 100]), title=None)
)


df_a_match['x'] = [pos[0]['x'] for pos in df_a_match['positions']]
df_a_match['y'] = [pos[0]['y'] for pos in df_a_match['positions']]
df_a_match.drop('positions', axis=1, inplace=True)
df_a_match.drop('tags', axis=1, inplace=True)

# brush2 = alt.selection_interval(
#     on="[mousedown[event.shiftKey], mouseup] > mousemove",
#     translate="[mousedown[event.shiftKey], mouseup] > mousemove!",
# )

brush2 = alt.selection(type='interval', encodings=['x', 'y'])

team_event = alt.Chart(df_a_match).mark_point(
    size=50,
    opacity=1,
    filled=True
).encode(
    x=alt.X('x:Q', axis=alt.Axis(labels=False, ticks=False, grid=False)),
    y=alt.Y('y:Q', axis=alt.Axis(labels=False, ticks=False, grid=False)),
    color=alt.condition(brush2,
                        alt.Color('teamId:N', legend=None, scale=alt.Scale(domain=list(df_a_match['teamId'].unique()), range=['black', 'cyan'])),
                        alt.value('lightgray')), 
    tooltip=['eventName:N', 'teamId:N', 'x:Q', 'y:Q']
).add_selection(
    brush2
)


zoom = alt.selection_interval(
    bind='scales',
    on="[mousedown[!event.shiftKey], mouseup] > mousemove",
    translate="[mousedown[!event.shiftKey], mouseup] > mousemove!",
)



soccer_pitch = alt.layer(background, border_lines, midline, lines, white_dots, arc1, arc2, arc3, arc4, center_circle, team_event).properties(
    width=700,
    height=440,
    title="Lazio - Internazionale, 2 - 3"
)


soccer_pitch = soccer_pitch.add_selection(
    zoom,
)



bars = alt.Chart(df_a_match).mark_bar().encode(
    y=alt.Y('eventName:N', sort='-x', title=None),
    x=alt.X('count():Q', title='frequency'),
    color=alt.Color('eventName:N',legend=None)
).transform_filter(
   brush2
)


annotations_df = pd.DataFrame({
    'text': ['You can select part of the graph to see distributoon of events'],
    'x': [0],
    'y': [0]
})

annotations_chart = alt.Chart(annotations_df).mark_text(
    align='left',
    baseline='middle',
    fontSize=12,
    fontStyle='italic'
).encode(
    text='text:N'
).properties(
    width=700,
    height=20
)

soccer_pitch_with_annotations = alt.vconcat(
    soccer_pitch,
    annotations_chart,
    spacing=5
)


combined_chart2 = alt.hconcat(
    soccer_pitch_with_annotations,
    bars,
    spacing=10
).resolve_scale(
    color='independent'
)

combined_chart2.display()

# drawing the fifth chart
combined_chart = alt.vconcat(
    combined_chart1,
    combined_chart2,
    spacing=10
).resolve_scale(
    color='independent'
)
combined_chart

description1 = pn.pane.Markdown("Our Scientific Visualization Project is designed to provide a comprehensive overview of the process undertaken to reimagine the visual representations based on the scientific study. The visualizations selected for redesign include a bar chart detailing the percentage of different event types occurring within a soccer match and a histogram showcasing the distribution of event frequencies, alongside a spatial plot representing the locations of match events on a soccer field. Each graphic serves a distinct purpose: the bar chart allows for the quick comparison of event prevalence, the histogram provides an understanding of the variability and central tendency of event counts, and the spatial plot conveys the common areas of activity during a match.")
decription2 = pn.pane.Markdown("The first chart below is a bar chart detailing the percentage of different event types occurring within a soccer match.")
chart1_panel = pn.pane.Vega(chart1, sizing_mode='stretch_width')
decription3 = pn.pane.Markdown("The second chart below is a histogram showcasing the distribution of event frequencies.")
chart2_panel = pn.pane.Vega(chart2, sizing_mode='stretch_width')
decription4 = pn.pane.Markdown("The third chart below is a spatial plot representing the locations of match events on a soccer field. We made a connection between this graph and the first graph. People can select the event points here with zooming, the right is the interactive bar chart that shows the distribution of event frequency.")
combined_chart1_panel = pn.pane.Vega(combined_chart1, sizing_mode='stretch_width')
description5 = pn.pane.Markdown("We combined all the graphs we have and the chart below is what the whole things look like.")
combined_chart2_panel = pn.pane.Vega(combined_chart2, sizing_mode='stretch_width')
combined_chart_panel = pn.pane.Vega(combined_chart, sizing_mode='stretch_width')


dashboard = pn.Column(
    "# Scientific Visualization Project",  
    description1,
    decription2,
    chart1_panel, 
    decription3,
    chart2_panel,
    decription4,
    description5,
    combined_chart_panel 
)

dashboard.servable(title='Scientific Visualization Project')