import streamlit as st import pandas as pd import plotly.express as px import pycountry import plotly.graph_objects as go import plotly.figure_factory as ff import pycountry_convert as pc import pandas as pd import numpy as np import pycountry import pycountry_convert as pc user = pd.read_csv('train_users_2.csv') user = user[user['age'] < 100] user["year"] = user["date_account_created"].str[:4].astype(int) user["month"] = user["date_account_created"].str[5:7].astype(int) user['year-month'] = user['date_account_created'].str[:7] user['date_first_booking'] = user['date_first_booking'].replace(np.nan, '2020-13-31') user['month_booking'] = user['date_first_booking'].str[5:7].astype(int) user['year_booking'] = user['date_first_booking'].str[:4].astype(int) user['year-month_booking'] = user['date_first_booking'].str[:7] user["language"] = user["language"].str.upper() df = user external_stylesheets = ['https://codepen.io/chriddyp/pen/bWLwgP.css'] st.set_page_config(layout="wide") st.title("Airbnb New User Bookings") col0, col1, col2 = st.columns([0.5, 2, 2]) # Adjust the width ratios here with col0: region = st.selectbox("Select Region", df['language'].unique(), index=0) column = st.selectbox("Select Column", ['first_device_type', 'first_browser','affiliate_provider'], index=0) year = st.slider("Select Year", df['year'].min(), df['year'].max(), df['year'].max()) with col1: # Update graph based on column and year selection filtered_df = df[(df['year'] == year)] visit = filtered_df[column].value_counts() booking = filtered_df[filtered_df['year_booking']==year][column].value_counts() counts = pd.merge(visit, booking, left_index=True, right_index=True).reset_index() counts.columns = ['first_device_type', 'visit', 'booking'] counts['rate'] = counts['booking'] / counts['visit'] counts_t = counts.melt(id_vars=['first_device_type'], value_vars=['visit', 'booking']) fig = px.bar(counts_t, x="first_device_type", y="value", color='variable') fig.update_layout(yaxis2=dict(overlaying='y', side='right', range=[0, 1])) fig.add_trace(go.Scatter(x=counts['first_device_type'], y=counts['rate'], mode='lines+markers', name='Conversion Rate', yaxis='y2')) fig.update_layout(height=250, margin={'l': 20, 'b': 50, 't': 10, 'r': 10}, hovermode='closest') st.plotly_chart(fig) # Update new-users figure based on region selection dff = df[df['language'] == region] visit = dff.groupby(['year', 'month'])['year-month'].value_counts().reset_index(name='count') visit = visit.iloc[:, 2:] booking = dff.groupby(['year_booking', 'month_booking'])['year-month_booking'].value_counts().reset_index(name='count') booking = booking[booking['year_booking'] != 2020] booking = booking.iloc[:, 2:] counts = pd.merge(visit, booking, left_on=['year-month'], right_on=['year-month_booking']) counts.columns = [ 'year-month', 'visit', 'year-month_booking', 'booking'] counts['rate'] = counts['booking'] / counts['visit'] fig = go.Figure() fig.add_trace(go.Scatter(x=counts['year-month'], y=counts['visit'], mode='lines', name='visit')) fig.add_trace(go.Scatter(x=counts['year-month'], y=counts['booking'], mode='lines', name='booking')) fig.update_layout(yaxis2=dict(overlaying='y', side='right', range=[0, 1])) fig.add_trace(go.Scatter(x=counts['year-month'], y=counts['rate'], mode='lines+markers', name='Conversion Rate', yaxis='y2')) fig.update_layout(height=250, margin={'l': 20, 'b': 30, 'r': 10, 't': 10}) st.plotly_chart(fig) with col2: # Update destination-country figure based on year selection dff = df.loc[(df["country_destination"]!="NDF") & (df["country_destination"]!="other")] dff = dff[dff['year'] == year] counts = dff['country_destination'].value_counts() counts = pd.DataFrame(counts) counts.reset_index(inplace=True) counts.columns = ['country_destination', 'count'] counts['country'] = counts['country_destination'].apply(lambda x: pycountry.countries.get(alpha_2=x).alpha_3) counts['continent'] = counts['country_destination'].apply(lambda x: pc.convert_continent_code_to_continent_name(pc.country_alpha2_to_continent_code(x))) fig = px.scatter_geo(counts, locations="country", color="continent", size='count', hover_name=counts['country']) fig.update_traces(customdata=counts['country']) title = '{}'.format(year) fig.add_annotation(x=0, y=0.85, xanchor='left', yanchor='bottom', xref='paper', yref='paper', showarrow=False, align='left', text=title) fig.update_layout(height=250, margin={'l': 20, 'b': 30, 'r': 10, 't': 10}) fig.update_layout(clickmode='event+select') st.plotly_chart(fig) # Update new-users-age figure based on selected country country_name = st.selectbox("Select Country", df['country_destination'].unique(), index=0) dff = df[df['country_destination'] == country_name] dff = dff[dff['year'] == year] # Modify this line to match your data dff = dff[dff['gender'] != '-unknown-'] fig = px.violin(dff, x="gender", y="age", box=True, color="gender", violinmode='overlay') title = '{}'.format(year) fig.add_annotation(x=0, y=0.85, xanchor='left', yanchor='bottom', xref='paper', yref='paper', showarrow=False, align='left', text=title) fig.update_layout(height=225, margin={'l': 20, 'b': 30, 'r': 10, 't': 10}) st.plotly_chart(fig)