import pandas as pd #chart import seaborn as sns import matplotlib.pyplot as plt #Map import folium import branca

sns.set(style='darkgrid') def get_label(g): for p in g.patches: height = p.get_height() g.text(p.get_x()+p.get_width()/2., height/2, '{}'.format(round(height)), ha="center", color='white')

jan = '/work/files/01.json' feb = '/work/files/02.json' mar = '/work/files/03.json' apr = '/work/files/04.json' may = '/work/files/05.json' jun = '/work/files/06.json'

js = [jan, feb, mar, apr, may, jun] df = pd.concat(map(pd.read_json, js))

df.tail()

df.shape

df.describe()

print('Number of rows {}'.format(df.shape[0])) print('Number of cols {}'.format(df.shape[1]))

df.sample()

# drop irrelevant column cause is a unique valeu to each trip cols = ['start_station_description', 'end_station_description'] df.drop(cols, axis=1, inplace=True)

df.isna().sum()

df.dtypes

df['started_at'] = pd.to_datetime(df['started_at']).dt.tz_localize(None) df['ended_at'] = pd.to_datetime(df['ended_at']).dt.tz_localize(None)

df.sample()

df1 = df.copy()

# Create a new column with name month of trip and weekday name month_map = {1: 'Jan', 2: 'Feb', 3: 'Mar', 4: 'Apr', 5: 'May', 6: 'Jun', 7: 'Jul', 8: 'Aug', 9: 'Sept', 10: 'Oct', 11: 'Nov', 12: 'Dec'} df1['month_trip'] = df1['started_at'].dt.month.map(month_map) day_week = month_map = {0: 'Monday', 1: 'Tuesday', 2: 'Wednesday', 3: 'Thursday', 4: 'Friday', 5: 'Saturday', 6: 'Sunday'} df1['day_week'] = df1['started_at'].dt.weekday.map(day_week)

#Create a new colum df1['route'] = df1[['start_station_name', 'end_station_name']].agg(' at '.join, axis=1)

df1.sample()

df2 = df1.copy()

#I will only do the analysis from January to May because we are still in June mask = (df2['month_trip'] != 'Jun') df_mask = df2.loc[mask]

month_trips = df_mask.groupby('month_trip', as_index=False)['route'].count().rename(columns={'route': 'total'}).sort_values('total', ascending=False) month_trips.head()

plt.figure(figsize=[15,6]) g = sns.barplot(x='month_trip', y='total', data=month_trips); plt.xlabel("Month Trip") get_label(g)

total_day = df_mask.groupby('day_week', as_index=False)['route'].count().rename(columns={'route': 'total'}).sort_values('total', ascending=False)

plt.figure(figsize=[15,6]) g = sns.barplot(x='day_week', y='total', data=total_day); plt.xlabel("Day of week") get_label(g)

df_mask.sample()

total_start = df_mask.groupby('start_station_name', as_index=False)['route'].count().rename(columns={'route': 'total'}).sort_values('total', ascending=False)

mask_AM = (df_mask['started_at'].dt.hour < 12) mask_PM = (df_mask['started_at'].dt.hour >= 12) df_am = df_mask.loc[mask_AM] df_pm = df_mask.loc[mask_PM]

plt.figure(figsize=[20,6]) g = sns.countplot(x=df_mask['started_at'].dt.hour, data=df_mask);

plt.figure(figsize=[20,6]) g = sns.countplot(x=df_am['started_at'].dt.hour, hue='day_week', data=df_am);

plt.figure(figsize=[20,6]) sns.countplot(x=df_pm['started_at'].dt.hour, hue='day_week', data=df_pm);

df_mask.sample()

started_station = df_mask.groupby('start_station_name', as_index=False)['start_station_id'].count().rename(columns={'start_station_id': 'total'}).sort_values('total', ascending=False).head(10) ended_station = df_mask.groupby('end_station_name', as_index=False)['end_station_id'].count().rename(columns={'end_station_id': 'total'}).sort_values('total', ascending=False).head(10) route = df_mask.groupby('route', as_index=False)['start_station_id'].count().rename(columns={'start_station_id': 'total'}).sort_values('total', ascending=False).head(10)

plt.figure(figsize=[20,6]) g = sns.barplot(x='start_station_name', y='total', data=started_station); plt.xlabel("Start Station") get_label(g)

plt.figure(figsize=[20,6]) g = sns.barplot(x='end_station_name', y='total', data=ended_station); plt.xlabel("End Station") get_label(g)

df_mask.sample()

plt.figure(figsize=[18,10]) g = sns.barplot(x='total', y='route', data=route); plt.yticks(fontsize=20) plt.xticks(fontsize=20) plt.title('MOST FREQUENT ROUTE');

aux = df_mask.groupby(['start_station_name', 'start_station_latitude', 'start_station_longitude'], as_index=False)['route'].count().rename(columns={'route': 'total'}).sort_values('total', ascending=False) aux = pd.DataFrame(aux) aux1 = df_mask.groupby(['end_station_name', 'end_station_latitude', 'end_station_longitude'], as_index=False)['route'].count().rename(columns={'route': 'total'}).sort_values('total', ascending=False) aux1 = pd.DataFrame(aux1) aux1.rename(columns={'total': 'end_total', 'end_station_name': 'start_station_name'}, inplace=True) aux1.drop(['end_station_latitude', 'end_station_longitude'], axis=1, inplace=True) aux3 = pd.merge(aux, aux1, how = 'inner', on = 'start_station_name') aux3.rename(columns={'start_station_name': 'Stations', 'start_station_latitude': 'latitude', 'start_station_longitude': 'longitude', 'total': 'start_total'}, inplace=True) aux3.head()

#location is the mean of every lat and long point to centre the map. location = aux3['latitude'].mean(), aux3['longitude'].mean() map = folium.Map(location=location,zoom_start=14, width='100%',height=600) for i, row in aux3.iterrows(): html=f""" <h3> Travel information </h3> <strong>Station: {row['Stations']}<br><br>Started Trips: { row['start_total']}<br><br>Ended Trips: {row['end_total']}</strong> """ iframe = folium.IFrame(html=html, width=260, height=190) popup = folium.Popup(iframe, max_width=2650) folium.Marker( location=[row['latitude'], row['longitude']], icon=folium.features.CustomIcon('https://cdn3.iconfinder.com/data/icons/flat-icons-web/40/Bicycle-512.png',icon_size=(50,50)), popup = popup, tooltip="Click for more information", ).add_to(map)