Code Editor for app.py

import streamlit as st
import pandas as pd
import numpy as np
import pydeck as pdk
from datetime import datetime, timedelta
import plotly.express as px

# Set page config
st.set_page_config(layout="wide", page_title="NYC Interactive Map")

# Initialize session state
if "map_style" not in st.session_state:
    st.session_state.map_style = "mapbox://styles/mapbox/streets-v11"
if "show_trucks" not in st.session_state:
    st.session_state.show_trucks = True
if "show_heatmap" not in st.session_state:
    st.session_state.show_heatmap = True
if "show_routes" not in st.session_state:
    st.session_state.show_routes = False
if "show_taxis" not in st.session_state:
    st.session_state.show_taxis = False

# Generate mock data
@st.cache_data
def generate_mock_data():
    # Food truck locations
    np.random.seed(42)
    n_trucks = 100
    trucks_df = pd.DataFrame({
        'latitude': np.random.normal(40.7128, 0.1, n_trucks),
        'longitude': np.random.normal(-74.0060, 0.1, n_trucks),
        'name': [f"Food Truck #{i}" for i in range(n_trucks)],
        'type': np.random.choice(['Hot Dogs', 'Tacos', 'Halal', 'Ice Cream'], n_trucks),
        'rating': np.random.uniform(3.0, 5.0, n_trucks),
        'borough': np.random.choice(['Manhattan', 'Brooklyn', 'Queens', 'Bronx', 'Staten Island'], n_trucks)
    })

    # Pedestrian activity (heatmap)
    n_points = 1000
    heatmap_df = pd.DataFrame({
        'latitude': np.random.normal(40.7128, 0.15, n_points),
        'longitude': np.random.normal(-74.0060, 0.15, n_points),
        'weight': np.random.exponential(1, n_points)
    })

    return trucks_df, heatmap_df

trucks_df, heatmap_df = generate_mock_data()

# Sidebar controls
st.sidebar.title("Map Controls")

# Map style selector
map_style = st.sidebar.selectbox(
    "Map Style",
    ["Streets", "Satellite"],
    key="map_style_selector"
)
st.session_state.map_style = "mapbox://styles/mapbox/satellite-v9" if map_style == "Satellite" else "mapbox://styles/mapbox/streets-v11"

# Layer toggles
st.sidebar.subheader("Layers")
st.session_state.show_trucks = st.sidebar.checkbox("Food Trucks", value=True)
st.session_state.show_heatmap = st.sidebar.checkbox("Pedestrian Heatmap", value=True)
st.session_state.show_routes = st.sidebar.checkbox("Animated Routes", value=False)
st.session_state.show_taxis = st.sidebar.checkbox("Taxi Clusters", value=False)

# Borough filter
selected_boroughs = st.sidebar.multiselect(
    "Filter by Borough",
    options=trucks_df['borough'].unique(),
    default=trucks_df['borough'].unique()
)

# Filter data
filtered_trucks = trucks_df[trucks_df['borough'].isin(selected_boroughs)]

# Create map layers
layers = []

if st.session_state.show_trucks:
    trucks_layer = pdk.Layer(
        "ScatterplotLayer",
        filtered_trucks,
        get_position=['longitude', 'latitude'],
        get_color=[255, 0, 0, 160],
        get_radius=100,
        pickable=True,
    )
    layers.append(trucks_layer)

if st.session_state.show_heatmap:
    heatmap_layer = pdk.Layer(
        "HeatmapLayer",
        heatmap_df,
        get_position=['longitude', 'latitude'],
        get_weight='weight',
        opacity=0.6,
    )
    layers.append(heatmap_layer)

# Create the map
st.title("NYC Interactive Map")

col1, col2 = st.columns([3, 1])

with col1:
    map_view = pdk.ViewState(
        latitude=40.7128,
        longitude=-74.0060,
        zoom=11,
        pitch=0
    )

    deck = pdk.Deck(
        map_style=st.session_state.map_style,
        initial_view_state=map_view,
        layers=layers,
        tooltip={
            "html": "<b>Food Truck:</b> {name}<br/>"
                   "<b>Type:</b> {type}<br/>"
                   "<b>Rating:</b> {rating:.1f}⭐<br/>"
                   "<b>Borough:</b> {borough}",
            "style": {
                "backgroundColor": "white",
                "color": "black"
            }
        }
    )

    st.pydeck_chart(deck)

# Stats panel
with col2:
    st.subheader("Statistics")
    
    # Trucks by borough
    st.write("Food Trucks by Borough")
    borough_counts = filtered_trucks['borough'].value_counts()
    fig = px.pie(values=borough_counts.values, names=borough_counts.index, hole=0.3)
    st.plotly_chart(fig, use_container_width=True)
    
    # Average ratings
    st.write("Average Ratings by Type")
    avg_ratings = filtered_trucks.groupby('type')['rating'].mean().sort_values(ascending=True)
    fig = px.bar(avg_ratings)
    st.plotly_chart(fig, use_container_width=True)

    # Summary stats
    st.metric("Total Food Trucks", len(filtered_trucks))
    st.metric("Average Rating", f"{filtered_trucks['rating'].mean():.1f}⭐")