import pandas as pd

import matplotlib.pyplot as plt

import seaborn as sns

import streamlit as st

st.set_page_config(layout="wide")

def create_monthly_avg_pollutant(df, year):

df['year_month'] = pd.to_datetime(df['year_month'])

monthly_avg_pollutant = df.groupby('year_month')[['SO2', 'NO2', 'O3']].mean().reset_index()

monthly_avg_pollutant = monthly_avg_pollutant[monthly_avg_pollutant['year_month'].dt.year==year]

return monthly_avg_pollutant

def create_monthly_avg_pm25(df):

df['year_month'] = pd.to_datetime(df['year_month'])

pm25_by_station_year = df.groupby(by=['station', 'year_month']).agg(avg_pm25=('PM2.5', 'mean')).reset_index()

pm25_by_station_year = pm25_by_station_year[pm25_by_station_year['year_month'].dt.year > 2013]

pm25_by_station_year = pm25_by_station_year[pm25_by_station_year['year_month'].dt.year < 2017]

return pm25_by_station_year

def getHighestandLowest(df, range):

df = df[df['year_month'].dt.year >= range[0]]

df = df[df['year_month'].dt.year <= range[1]]

sorted_df = df.groupby(by="station").agg({

'PM2.5':['mean'],

}).sort_values(by=("PM2.5","mean"), ascending=False).reset_index()

return sorted_df["station"].iloc[0], sorted_df["station"].iloc[-1]

all_df = pd.read_csv("all_data.csv")

aoti_df = all_df[all_df["station"]=="Aotizhongxin"]

changping_df = all_df[all_df["station"]=="Changping"]

dingling_df = all_df[all_df["station"]=="Dingling"]

dongsi_df = all_df[all_df["station"]=="Dongsi"]

guanyuan_df = all_df[all_df["station"]=="Guanyuan"]

gucheng_df = all_df[all_df["station"]=="Gucheng"]

huairou_df = all_df[all_df["station"]=="Huairou"]

nongzhan_df = all_df[all_df["station"]=="Nongzhanguan"]

shunyi_df = all_df[all_df["station"]=="Shunyi"]

tiantan_df = all_df[all_df["station"]=="Tiantan"]

wanliu_df = all_df[all_df["station"]=="Wanliu"]

wanshou_df = all_df[all_df["station"]=="Wanshou"]

st.header('Air Quality Analysis :sparkles:')

st.subheader('')

st.subheader("Average SO2, NO2 and O3 Statistics")

col1, col2 = st.columns(2)

with col1:

station = st.selectbox(

label="Select station for monthly analysis",

options=('Aotizhongxin', 'Changping', 'Dingling', 'Guanyuan', 'Gucheng', 'Huairou', 'Nongzhanguan', 'Shunyi', 'Tiantan', 'Wanliu', 'Wanshou')

)

match station:

case "Aotizhongxin":

selected_df = aoti_df

case "Changping":

selected_df = changping_df

case "Dingling":

selected_df = dingling_df

case "Guanyuan":

selected_df = guanyuan_df

case "Gucheng":

selected_df = gucheng_df

case "Huairou":

selected_df = huairou_df

case "Nongzhanguan":

selected_df = nongzhan_df

case "Shunyi":

selected_df = shunyi_df

case "Tiantan":

selected_df = tiantan_df

case "Wanliu":

selected_df = wanliu_df

case "Wanshouxigong":

selected_df = wanshou_df

year = st.radio(

label="Select year",

options=(2014, 2015, 2016),

horizontal=False

)

monthly_avg_pollutant = create_monthly_avg_pollutant(selected_df, year)

with col2:

fig, ax = plt.subplots(figsize=(10,6))

ax.plot(monthly_avg_pollutant["year_month"], monthly_avg_pollutant["SO2"], label="SO2", marker='o')

ax.plot(monthly_avg_pollutant["year_month"], monthly_avg_pollutant["NO2"], label="NO2", marker='o')

ax.plot(monthly_avg_pollutant["year_month"], monthly_avg_pollutant["O3"], label="O3", marker='o')

plt.title(f"Average SO2, NO2, and O3 per Month ({year}) in {station}")

plt.grid(True)

ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))

st.pyplot(fig)

st.subheader("Statistics of Average PM2.5")

col1, col2 = st.columns(2)

with col1:

values = st.slider(

label='Select a range of year',

min_value=2014, max_value=2016, value=(2014, 2016))

pm25_by_station_year = create_monthly_avg_pm25(all_df)

pm25_by_station_year = pm25_by_station_year[pm25_by_station_year['year_month'].dt.year>=values[0]]

pm25_by_station_year = pm25_by_station_year[pm25_by_station_year['year_month'].dt.year<=values[1]]

df_station = pm25_by_station_year.groupby(by=['station'])

highest, lowest = getHighestandLowest(all_df, values)

st.metric("Highes PM2.5 Average", value=highest)

st.metric("Lowest PM2.5 Average", value=lowest)

with col2:

fig, ax = plt.subplots(figsize=(11,8))

custom_colors = {highest: 'red', lowest: 'blue'}

for key, group in df_station:

if key[0] in custom_colors:

color = custom_colors[key[0]]

opacity = 1

else:

color = 'gray'

opacity = 0.4

ax.plot(group['year_month'], group['avg_pm25'], label=key, color=color, alpha=opacity, marker="o")

ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))

plt.grid()

plt.xticks(rotation=45)

plt.title(f"Average PM2.5 per Month ({values[0]}-{values[1]}) in Every District" if values[0]<values[1] else f"Average PM2.5 per Month ({values[0]}) in Every District")

st.pyplot(fig)