Spaces:
Sleeping
Sleeping
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,148 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
import pandas as pd
|
3 |
+
import plotly.express as px
|
4 |
+
import plotly.graph_objects as go
|
5 |
+
import numpy as np
|
6 |
+
|
7 |
+
def load_and_preprocess_data(file_path):
|
8 |
+
# Read the data
|
9 |
+
df = pd.read_csv(file_path)
|
10 |
+
|
11 |
+
# Drop redundant columns
|
12 |
+
df = df.drop(['X', 'Y'], axis=1)
|
13 |
+
|
14 |
+
# Handle missing values
|
15 |
+
df.dropna(subset=['Incidentid', 'DateTime', 'Year', 'Latitude', 'Longitude'], inplace=True)
|
16 |
+
|
17 |
+
# Fill numeric values
|
18 |
+
numeric = ['Age_Drv1', 'Age_Drv2']
|
19 |
+
for col in numeric:
|
20 |
+
df[col].fillna(df[col].median(), inplace=True)
|
21 |
+
|
22 |
+
# Fill categorical values
|
23 |
+
categorical = [
|
24 |
+
'Gender_Drv1', 'Violation1_Drv1', 'AlcoholUse_Drv1', 'DrugUse_Drv1',
|
25 |
+
'Gender_Drv2', 'Violation1_Drv2', 'AlcoholUse_Drv2', 'DrugUse_Drv2',
|
26 |
+
'Unittype_Two', 'Traveldirection_Two', 'Unitaction_Two', 'CrossStreet'
|
27 |
+
]
|
28 |
+
for col in categorical:
|
29 |
+
df[col].fillna('Unknown', inplace=True)
|
30 |
+
|
31 |
+
# Remove invalid ages
|
32 |
+
df = df[
|
33 |
+
(df['Age_Drv1'] <= 90) &
|
34 |
+
(df['Age_Drv2'] <= 90) &
|
35 |
+
(df['Age_Drv1'] >= 16) &
|
36 |
+
(df['Age_Drv2'] >= 16)
|
37 |
+
]
|
38 |
+
|
39 |
+
# Create age groups for both drivers
|
40 |
+
df['Age_Group_Drv1'] = pd.cut(
|
41 |
+
df['Age_Drv1'],
|
42 |
+
bins=[15, 25, 35, 45, 55, 65, 90],
|
43 |
+
labels=['16-25', '26-35', '36-45', '46-55', '56-65', '65+']
|
44 |
+
)
|
45 |
+
|
46 |
+
df['Age_Group_Drv2'] = pd.cut(
|
47 |
+
df['Age_Drv2'],
|
48 |
+
bins=[15, 25, 35, 45, 55, 65, 90],
|
49 |
+
labels=['16-25', '26-35', '36-45', '46-55', '56-65', '65+']
|
50 |
+
)
|
51 |
+
|
52 |
+
return df
|
53 |
+
|
54 |
+
def create_severity_violation_chart(df, selected_age_group=None):
|
55 |
+
# Filter by age group if selected
|
56 |
+
if selected_age_group:
|
57 |
+
df = df[
|
58 |
+
(df['Age_Group_Drv1'] == selected_age_group) |
|
59 |
+
(df['Age_Group_Drv2'] == selected_age_group)
|
60 |
+
]
|
61 |
+
|
62 |
+
# Create violation categories for both drivers
|
63 |
+
violations_drv1 = df.groupby(['Violation1_Drv1', 'Injuryseverity']).size().reset_index(name='count')
|
64 |
+
violations_drv2 = df.groupby(['Violation1_Drv2', 'Injuryseverity']).size().reset_index(name='count')
|
65 |
+
|
66 |
+
# Combine violations from both drivers
|
67 |
+
violations_drv1.columns = ['Violation', 'Severity', 'count']
|
68 |
+
violations_drv2.columns = ['Violation', 'Severity', 'count']
|
69 |
+
violations_combined = pd.concat([violations_drv1, violations_drv2])
|
70 |
+
|
71 |
+
# Aggregate the combined violations
|
72 |
+
violations_agg = violations_combined.groupby(['Violation', 'Severity'])['count'].sum().reset_index()
|
73 |
+
|
74 |
+
# Create the stacked bar chart
|
75 |
+
fig = px.bar(
|
76 |
+
violations_agg,
|
77 |
+
x='Violation',
|
78 |
+
y='count',
|
79 |
+
color='Severity',
|
80 |
+
title=f'Distribution of Crash Severity by Violation Type {selected_age_group if selected_age_group else ""}',
|
81 |
+
labels={'count': 'Number of Incidents', 'Violation': 'Violation Type'},
|
82 |
+
height=600
|
83 |
+
)
|
84 |
+
|
85 |
+
# Customize the layout
|
86 |
+
fig.update_layout(
|
87 |
+
xaxis_tickangle=-45,
|
88 |
+
legend_title='Severity',
|
89 |
+
barmode='stack',
|
90 |
+
showlegend=True
|
91 |
+
)
|
92 |
+
|
93 |
+
return fig
|
94 |
+
|
95 |
+
def main():
|
96 |
+
st.title('Traffic Crash Analysis Dashboard')
|
97 |
+
|
98 |
+
# Load data
|
99 |
+
df = load_and_preprocess_data('1.08_Crash_Data_Report_(detail).csv')
|
100 |
+
|
101 |
+
# Create age group selector
|
102 |
+
st.sidebar.header('Filters')
|
103 |
+
age_groups = ['All'] + list(df['Age_Group_Drv1'].unique())
|
104 |
+
selected_age_group = st.sidebar.selectbox('Select Age Group', age_groups)
|
105 |
+
|
106 |
+
# Create and display the chart
|
107 |
+
if selected_age_group == 'All':
|
108 |
+
fig = create_severity_violation_chart(df)
|
109 |
+
else:
|
110 |
+
fig = create_severity_violation_chart(df, selected_age_group)
|
111 |
+
|
112 |
+
st.plotly_chart(fig, use_container_width=True)
|
113 |
+
|
114 |
+
# Add additional insights
|
115 |
+
st.subheader('Analysis Insights')
|
116 |
+
|
117 |
+
# Calculate and display some statistics
|
118 |
+
if selected_age_group == 'All':
|
119 |
+
total_crashes = len(df)
|
120 |
+
else:
|
121 |
+
total_crashes = len(df[
|
122 |
+
(df['Age_Group_Drv1'] == selected_age_group) |
|
123 |
+
(df['Age_Group_Drv2'] == selected_age_group)
|
124 |
+
])
|
125 |
+
|
126 |
+
st.write(f"Total number of crashes: {total_crashes:,}")
|
127 |
+
|
128 |
+
# Show top violations
|
129 |
+
st.subheader('Top Violations')
|
130 |
+
if selected_age_group == 'All':
|
131 |
+
violations = pd.concat([
|
132 |
+
df['Violation1_Drv1'].value_counts(),
|
133 |
+
df['Violation1_Drv2'].value_counts()
|
134 |
+
]).groupby(level=0).sum()
|
135 |
+
else:
|
136 |
+
filtered_df = df[
|
137 |
+
(df['Age_Group_Drv1'] == selected_age_group) |
|
138 |
+
(df['Age_Group_Drv2'] == selected_age_group)
|
139 |
+
]
|
140 |
+
violations = pd.concat([
|
141 |
+
filtered_df['Violation1_Drv1'].value_counts(),
|
142 |
+
filtered_df['Violation1_Drv2'].value_counts()
|
143 |
+
]).groupby(level=0).sum()
|
144 |
+
|
145 |
+
st.write(violations.head())
|
146 |
+
|
147 |
+
if __name__ == "__main__":
|
148 |
+
main()
|