def armed_or_not(self, img_out=False, csv_out=False, path=None):
arm_df = self.shoots_df.select(c.ARMED, 'race_name')\
.na.drop(subset=[c.ARMED])\
.na.drop(subset=['race_name'])
arm_df = arm_df.withColumn(c.ARMED, udf_convert_armed(c.ARMED))\
.groupby('race_name', c.ARMED)\
.count()
for label in ['armed', 'unarmed']:
_arm_df = arm_df.filter(col(c.ARMED) == label)
arm_dfpd = self_toPandas(_arm_df)
if csv_out:
self._save_csv(arm_dfpd, f'{path}/{label}.csv')
if img_out:
fig = go.Figure(data=[
go.Pie(labels=arm_dfpd['race_name'],
values=arm_dfpd['count'],
hole=.3,
textinfo='label+percent',
insidetextorientation='radial')
])
fig.update_layout(title_text=f'USA: {label}', title_x=0.5)
fig.write_image(f'{path}/{label}.png')
def crimes_per_state(self, img_out=False, csv_out=False, path=None):
shoot_state = self.shoots_df.select(c.STATE)\
.groupby(c.STATE)\
.count()\
.sort(col('count'))
shoot_state = self_toPandas(shoot_state)
if csv_out:
self._save_csv(shoot_state, f'{path}/crimes_per_states.csv')
if img_out:
fig = go.Figure(
go.Bar(
y=shoot_state[c.STATE],
x=shoot_state['count'],
orientation='h',
text=shoot_state['count'],
textposition='outside',
marker_color=shoot_state['count'],
))
fig.update_xaxes(showline=True,
linewidth=2,
linecolor='black',
mirror=True)
fig.update_yaxes(showline=True,
linewidth=2,
linecolor='black',
mirror=True)
fig.update_layout(title_text='Fatal Killing - All States',
yaxis_title='States',
xaxis_title='Total number of victims',
title_x=0.5,
height=1000)
fig.write_image(f'{path}/crimes_per_states.png')
def kills_per_year(self, img_out=False, csv_out=False, path=None):
year_shoot = self.shoots_df.select('year')\
.groupby('year')\
.count()\
.sort(col('year'))
year_shoot = self_toPandas(year_shoot)
if csv_out:
self._save_csv(year_shoot, f'{path}/kills_year_sp.csv')
if img_out:
fig = go.Figure(data=go.Scatter(x=year_shoot['year'],
y=year_shoot['count'],
mode='lines+markers',
marker_color='red'))
fig.update_xaxes(showline=True,
linewidth=2,
linecolor='black',
mirror=True)
fig.update_yaxes(showline=True,
linewidth=2,
linecolor='black',
mirror=True)
fig.update_layout(title_text='Deaths - All Years',
xaxis_title='Years',
yaxis_title='Total number of kills',
title_x=0.5)
fig.write_image(f'{path}/kills_year_sp.png')
def flee(self, img_out=False, csv_out=False, path=None):
arm_df = self.shoots_df.select(c.FLEE).groupby(c.FLEE).count()
arm_df = self_toPandas(arm_df)
if csv_out:
self._save_csv(arm_df, f'{path}/flee.csv')
if img_out:
fig = go.Figure(data=[
go.Pie(labels=arm_df['flee'],
values=arm_df['count'],
hole=.3,
textinfo='label+percent',
insidetextorientation='radial')
])
fig.update_layout(title_text='Victim Flee?', title_x=0.5)
fig.write_image(f'{path}/flee.png')
def monthly(self, img_out=False, csv_out=False, path=None):
monthly_df = self.shoots_df.select(c.DATE)\
.withColumn(c.DATE, udf_convert_date_to_string(c.DATE))\
.groupby(c.DATE)\
.count()\
.sort(col(c.DATE))
monthly_df = self_toPandas(monthly_df)
month_year = [str(i) for i in monthly_df[c.DATE]]
if csv_out:
self._save_csv(monthly_df, f'{path}/monthlysp.csv')
if img_out:
fig = make_subplots(
rows=2,
cols=1,
subplot_titles=('Monthly series',
'Distribution of monthly count'))
fig.add_trace(go.Scatter(
x=month_year,
y=monthly_df['count'],
name='Monthly Deaths',
mode='lines',
),
row=1,
col=1)
fig.add_trace(go.Box(y=monthly_df['count'],
name='Count',
marker_color='indianred',
boxmean='sd'),
row=2,
col=1)
fig.update_xaxes(title_text='Year',
row=1,
col=1,
showline=True,
linewidth=2,
linecolor='black',
mirror=True)
fig.update_xaxes(title_text=' ',
row=2,
col=1,
showline=True,
linewidth=2,
linecolor='black',
mirror=True)
fig.update_yaxes(title_text='Number of Victims',
row=1,
col=1,
showline=True,
linewidth=2,
linecolor='black',
mirror=True)
fig.update_yaxes(title_text='Number of Victims',
row=2,
col=1,
showline=True,
linewidth=2,
linecolor='black',
mirror=True)
fig.update_layout(
title_text='Fatal Killing Monthly Count 2015 - 2020',
title_x=0.5,
showlegend=False,
height=1000)
fig.write_image(f'{path}/monthlysp.png')
return monthly_df
def races(self, img_out=False, csv_out=False, path=None):
shoot_race = self.shoots_df.groupby(['year', 'month_num',
'race_name']).count()
shoot_race = self_toPandas(shoot_race)
shoot_race['monthly'] = shoot_race['year'].astype(
str) + '-' + shoot_race['month_num'].astype(str)
if csv_out:
self._save_csv(shoot_race, f'{path}/racetoll.csv')
if img_out:
fig = make_subplots(rows=3,
cols=2,
subplot_titles=('Black', 'White', 'Hispanic',
'Asian', 'Native American',
'Others'))
fig.add_trace(self.plot_month_race(shoot_race, 'Black', 'brown'),
row=1,
col=1)
fig.add_trace(self.plot_month_race(shoot_race, 'White',
'deepskyblue'),
row=1,
col=2)
fig.add_trace(self.plot_month_race(shoot_race, 'Hispanic',
'green'),
row=2,
col=1)
fig.add_trace(self.plot_month_race(shoot_race, 'Asian', 'red'),
row=2,
col=2)
fig.add_trace(self.plot_month_race(shoot_race, 'Native American',
'orange'),
row=3,
col=1)
fig.add_trace(self.plot_month_race(shoot_race, 'Others', 'violet'),
row=3,
col=2)
fig.update_layout(title_text='Deaths - All Race', title_x=0.5)
fig.write_image(f'{path}/affrace.png')
fig = go.Figure()
races = [
'Black', 'White', 'Hispanic', 'Asian', 'Native American',
'Others'
]
colors = ['brown', 'grey', 'green', 'red', 'orange', 'violet']
for race, color in zip(races, colors):
fig.add_trace(
go.Box(y=shoot_race.loc[(
shoot_race['race_name'] == race)]['count'],
name=race,
marker_color=color,
boxmean=True))
fig.update_xaxes(showline=True,
linewidth=2,
linecolor='black',
mirror=True)
fig.update_yaxes(showline=True,
linewidth=2,
linecolor='black',
mirror=True)
fig.update_layout(title='Death Toll - All Races',
title_x=0.5,
xaxis=dict(title='Race'),
yaxis=dict(title='Number of Victims'))
fig.write_image(f'{path}/racetoll.png')