{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# ECON 323 Final Project\n", "\n", "###### Factors affecting bike share in San Francisco\n", "\n", "## Background \n", "Bike sharing is an increasingly popular service, especially as cities are moving to be more sustainable. [Eren and Uz (2020)](https://www.sciencedirect.com/science/article/abs/pii/S2210670719312387) identified several factors affecting bike sharing, including weather (e.g. seasons, precipitation, and temperature), public transportation, safety, and other temporal factors. Bay Wheels is a bike sharing service in San Francisco provided by Lyft (similar to the Shaw Mobi bikes in Vancouver). Some differences are that Bay Wheels provides both electric bikes and regular bikes, and that the service is accessible through the Lyft application and Clipper cards (similar to Vancouver’s Compass card). Bay Wheels’ dataset is publicly available on Lyft’s website at https://www.lyft.com/bikes/bay-wheels/system-data. This project aims to look at factors that influence bike sharing in San Francisco.\n", "\n", "## Literature Review\n", "Based on Eren and Uz (2020)’s findings, the optimal temperature for bike sharing hovers around 20 – 30 °C with no precipitation, while the 10-20 °C range also has a positive correlation in bike sharing. Eren and Uz also found that bike sharing can be used as a substitute for public transportation in areas with poor public transport options or at times where services are reduced (e.g. late night/early morning). However, bike sharing services are also used as complements to public transportation to reduce travel time. Eren and Uz found that the correlation between income level and usage of bike share services tend to be strongly correlated as income goes up. However, Guo et al (2017) found that in Ningbo, China found that the use of bike sharing services are highest among lower income residents, followed by middle-income and high-income respectively. \n", "\n", "## Hypothesis/Research\n", "\n", "Using Eren and Uz (2020)’s findings as a starting point, I will look at how the following factors affect bike sharing in San Francisco. \n", "\n", "* **Weather** – San Francisco’s mild coastal weather ranges around 8 – 21 °C all year round, with an average of 8 days of precipitation in the winter months and 0 days in the summer months. These numbers make San Francisco’s weather the ideal conditions for bike sharing. I will examine whether seasonality still influences bike sharing in San Francisco by cross-examining weather data with start/end times for bike share sessions.\n", "\n", "* **Public transportation** – I will look bike share usage along BART (rail/subway), Caltrain, and MUNI (light rail and cable car) stops and the time of day. Since many working in the city are commuters, I would expect higher usages along major transit stops before/after work hours. \n", "\n", "* **Safety** – I will see whether crime rate (from San Francisco Police Department) affects bike share usage. Safety is often cited as a concern for bike share users, and I expect there to be a negative correlation between bike share usage and crime rate. \n", "\n", "* **Temporal factors** - I will break down bike share usage by weekdays/weekends, and also the time of day. \n", "\n", "## Dataset\n", "The main dataset used in this analysis comes from Lyft's Bay Wheels bike share system data from June 2019 to March 2021 (the most recently available). It is available at https://www.lyft.com/bikes/bay-wheels/system-data. I have chosen to exclude data prior to Lyft's acquisitation of Bay Wheels in June 2019 (previously Ford GoBike), as the rebrand led to slight changes to the bike share model (e.g. docked and dockless bikes, increased electrical bikes, etc...) and the rental access methods (i.e. the bikes can now be rented through the Lyft app). Bay Wheels operates in San Francisco, Oakland, Berkeley, Emeryville, and San Jose. I will focus this study on bike sharing in San Francisco only. \n", "\n", "Other datasets used:\n", "* [San Francisco GeoPolygon](http://www2.census.gov/geo/tiger/PREVGENZ/ma/ma99/ce00shp/ce99_d00_shp.zip) for filtering for San Francisco data only\n", "* [Historical weather records from San Francisco Bay Area Weather Forecast Office](https://w2.weather.gov/climate/xmacis.php?wfo=mtr)\n", "* [Passenger rail stations data from California's Metropolitan Transportation Commission (MTC)](https://opendata.mtc.ca.gov/datasets/efd75b7bb3c04dbda06c6e7cd73e9336_0?geometry=-122.954%2C37.659%2C-121.976%2C37.849), which includes Bay Area Rapid Transit (BART), Caltrain, and San Francisco Municipal Railway (MUNI - light rail and cable car) data\n", "* [Incidents report from San Francisco Police Department (2018 to Present)](https://data.sfgov.org/Public-Safety/Police-Department-Incident-Reports-2018-to-Present/wg3w-h783)\n", "* [San Francisco Neighborhoods from San Francisco Open Data (DataSF)](https://data.sfgov.org/Geographic-Locations-and-Boundaries/Analysis-Neighborhoods/p5b7-5n3h)\n", "\n", "## Outline\n", "1. **Examining temporal factors:** visualizations of time of day and weekday/weekend usage of bike shares, as well as user types over weekday/weekends (casual versus subscribed member)\n", "\n", "2. **Examining the role of weather:** visualization of daily temperature, precipitation, and rate of bike shares\n", "\n", "3. **Examining safety and public transportation as factors affeting bike sharing:** map visualization of pubic transportation (BART, MUNI, Caltrain) stations, crime rate for each neighborhood, and bike share usage based on start location \n", "\n", "4. **Training a linear regression model to predict bike share usage** comparing linear regression and one with polynomial transformations applied\n", "\n", "5. **Conclusion:** discussion of limitations, areas for further research" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
started_atended_atstart_station_idstart_station_namestart_latstart_lngend_station_idend_station_nameend_latend_lngbike_idmember_casualbike_share_for_all_triprental_access_methodrideable_type
02019-06-30 18:16:09.77302019-07-01 16:57:45.592010917th St at Valencia St37.763316-122.42190456Koshland Park37.773414-122.4273171502.0memberNo<NA>NaN
12019-06-30 18:09:55.83002019-07-01 14:47:36.6810502nd St at Townsend St37.780526-122.39028810115th St at Potrero Ave37.767079-122.4073592526.0casualNo<NA>NaN
22019-06-30 15:40:31.03802019-07-01 08:13:54.349023The Embarcadero at Steuart St37.791464-122.39103430San Francisco Caltrain (Townsend St at 4th St)37.776598-122.3952822427.0memberNo<NA>NaN
42019-06-30 17:21:00.05502019-07-01 06:55:54.996015San Francisco Ferry Building (Harry Bridges Pl...37.795392-122.39420330San Francisco Caltrain (Townsend St at 4th St)37.776598-122.3952821070.0casualNo<NA>NaN
62019-06-30 14:31:39.57302019-07-01 00:53:02.25206The Embarcadero at Sansome St37.804770-122.403234400Buchanan St at North Point St37.804272-122.4335371980.0casualNo<NA>NaN
................................................
39449332021-03-17 15:28:272021-03-17 15:29:10SF-C28-2Broadway at Battery St37.798572-122.400869SF-C28-2Broadway at Battery St37.798572-122.400869NaNcasualNaNNaNclassic_bike
39449352021-03-26 11:27:132021-03-26 11:38:59SF-C28-2Broadway at Battery St37.798509-122.400834SF-C28-2Broadway at Battery St37.798637-122.400695NaNcasualNaNNaNelectric_bike
39449362021-03-21 11:24:062021-03-21 11:25:33SF-E18Divisadero St at Clay St37.789588-122.440683SF-E18Divisadero St at Clay St37.789588-122.440683NaNcasualNaNNaNclassic_bike
39449372021-03-29 20:34:272021-03-29 20:47:19SF-BB17London St at Geneva Ave37.716183-122.440112SF-BB17London St at Geneva Ave37.716209-122.440078NaNcasualNaNNaNelectric_bike
39449382021-03-15 18:18:242021-03-15 18:27:03SF-N27Rhode Island St at 17th St37.764478-122.402570SF-N27Rhode Island St at 17th St37.764478-122.402570NaNcasualNaNNaNclassic_bike
\n", "

3118591 rows × 15 columns

\n", "
" ], "text/plain": [ " started_at ended_at start_station_id \\\n", "0 2019-06-30 18:16:09.7730 2019-07-01 16:57:45.5920 109 \n", "1 2019-06-30 18:09:55.8300 2019-07-01 14:47:36.6810 50 \n", "2 2019-06-30 15:40:31.0380 2019-07-01 08:13:54.3490 23 \n", "4 2019-06-30 17:21:00.0550 2019-07-01 06:55:54.9960 15 \n", "6 2019-06-30 14:31:39.5730 2019-07-01 00:53:02.2520 6 \n", "... ... ... ... \n", "3944933 2021-03-17 15:28:27 2021-03-17 15:29:10 SF-C28-2 \n", "3944935 2021-03-26 11:27:13 2021-03-26 11:38:59 SF-C28-2 \n", "3944936 2021-03-21 11:24:06 2021-03-21 11:25:33 SF-E18 \n", "3944937 2021-03-29 20:34:27 2021-03-29 20:47:19 SF-BB17 \n", "3944938 2021-03-15 18:18:24 2021-03-15 18:27:03 SF-N27 \n", "\n", " start_station_name start_lat \\\n", "0 17th St at Valencia St 37.763316 \n", "1 2nd St at Townsend St 37.780526 \n", "2 The Embarcadero at Steuart St 37.791464 \n", "4 San Francisco Ferry Building (Harry Bridges Pl... 37.795392 \n", "6 The Embarcadero at Sansome St 37.804770 \n", "... ... ... \n", "3944933 Broadway at Battery St 37.798572 \n", "3944935 Broadway at Battery St 37.798509 \n", "3944936 Divisadero St at Clay St 37.789588 \n", "3944937 London St at Geneva Ave 37.716183 \n", "3944938 Rhode Island St at 17th St 37.764478 \n", "\n", " start_lng end_station_id \\\n", "0 -122.421904 56 \n", "1 -122.390288 101 \n", "2 -122.391034 30 \n", "4 -122.394203 30 \n", "6 -122.403234 400 \n", "... ... ... \n", "3944933 -122.400869 SF-C28-2 \n", "3944935 -122.400834 SF-C28-2 \n", "3944936 -122.440683 SF-E18 \n", "3944937 -122.440112 SF-BB17 \n", "3944938 -122.402570 SF-N27 \n", "\n", " end_station_name end_lat \\\n", "0 Koshland Park 37.773414 \n", "1 15th St at Potrero Ave 37.767079 \n", "2 San Francisco Caltrain (Townsend St at 4th St) 37.776598 \n", "4 San Francisco Caltrain (Townsend St at 4th St) 37.776598 \n", "6 Buchanan St at North Point St 37.804272 \n", "... ... ... \n", "3944933 Broadway at Battery St 37.798572 \n", "3944935 Broadway at Battery St 37.798637 \n", "3944936 Divisadero St at Clay St 37.789588 \n", "3944937 London St at Geneva Ave 37.716209 \n", "3944938 Rhode Island St at 17th St 37.764478 \n", "\n", " end_lng bike_id member_casual bike_share_for_all_trip \\\n", "0 -122.427317 1502.0 member No \n", "1 -122.407359 2526.0 casual No \n", "2 -122.395282 2427.0 member No \n", "4 -122.395282 1070.0 casual No \n", "6 -122.433537 1980.0 casual No \n", "... ... ... ... ... \n", "3944933 -122.400869 NaN casual NaN \n", "3944935 -122.400695 NaN casual NaN \n", "3944936 -122.440683 NaN casual NaN \n", "3944937 -122.440078 NaN casual NaN \n", "3944938 -122.402570 NaN casual NaN \n", "\n", " rental_access_method rideable_type \n", "0 NaN \n", "1 NaN \n", "2 NaN \n", "4 NaN \n", "6 NaN \n", "... ... ... \n", "3944933 NaN classic_bike \n", "3944935 NaN electric_bike \n", "3944936 NaN classic_bike \n", "3944937 NaN electric_bike \n", "3944938 NaN classic_bike \n", "\n", "[3118591 rows x 15 columns]" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import pandas as pd\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "from shapely.geometry import Point\n", "from geopandas import GeoDataFrame\n", "import geopandas as gpd\n", "\n", "all_files_to_format = ['201906-baywheels-tripdata.csv', '201907-baywheels-tripdata.csv',\n", " '201908-baywheels-tripdata.csv', '201909-baywheels-tripdata.csv', \n", " '201910-baywheels-tripdata.csv','201911-baywheels-tripdata.csv',\n", " '201912-baywheels-tripdata.csv', '202001-baywheels-tripdata.csv',\n", " '202002-baywheels-tripdata.csv', '202003-baywheels-tripdata.csv']\n", "\n", "all_files = ['202004-baywheels-tripdata.csv', '202005-baywheels-tripdata.csv', '202006-baywheels-tripdata.csv', \n", " '202007-baywheels-tripdata.csv', '202008-baywheels-tripdata.csv', '202009-baywheels-tripdata.csv', \n", " '202010-baywheels-tripdata.csv', '202011-baywheels-tripdata.csv', '202012-baywheels-tripdata.csv',\n", " '202101-baywheels-tripdata.csv', '202102-baywheels-tripdata.csv', '202103-baywheels-tripdata.csv']\n", "\n", "# read each month's data and append them to the df\n", "li = []\n", "for filename in all_files_to_format:\n", " df = pd.read_csv(filename, index_col=None, header=0, dtype={'bike_share_for_all_trip': 'string', 'rental_access_method': 'string'})\n", " # pre-2020 data has some discrepancies in column names and values, rename them to match 2020 data\n", " df = df.rename(columns={'start_station_latitude': 'start_lat', 'start_station_longitude': 'start_lng',\n", " 'end_station_latitude': 'end_lat', 'end_station_longitude': 'end_lng',\n", " 'user_type': 'member_casual', 'start_time': 'started_at', 'end_time': 'ended_at'\n", " })\n", " df = df.replace({'member_casual': r'Customer$'}, {'member_casual': 'casual'}, regex=True)\n", " df = df.replace({'member_casual': r'Subscriber$'}, {'member_casual': 'member'}, regex=True)\n", " li.append(df)\n", " \n", "for filename in all_files:\n", " df = pd.read_csv(filename, index_col=None, header=0, dtype={'bike_share_for_all_trip': 'string', 'rental_access_method': 'string'})\n", " li.append(df)\n", "\n", "lyft_data = pd.concat(li, axis=0, ignore_index=True)\n", "\n", "\n", "# add new column to df to filter for sf data only\n", "lyft_data['withinSF'] = 1\n", "withinSFlist = []\n", "\n", "# get sf polygon\n", "state_df = gpd.read_file(\"http://www2.census.gov/geo/tiger/PREVGENZ/ma/ma99/ce00shp/ce99_d00_shp.zip\")\n", "SF_polygon = state_df.loc[state_df['NAME'] == 'San Francisco'].iloc[1].geometry\n", "\n", "for lon,lat in zip(lyft_data.start_lng, lyft_data.start_lat):\n", " pt = Point(lon, lat)\n", " withinSF = SF_polygon.contains(pt)\n", " withinSFlist.append(withinSF)\n", " \n", "lyft_data['withinSF'] = withinSFlist\n", "lyft_data = lyft_data[lyft_data.withinSF == 1]\n", "\n", "# drop unused/renamed cols\n", "lyft_data = lyft_data.drop(columns=['withinSF', 'duration_sec', 'ride_id'])\n", "\n", "lyft_data" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
started_atended_atstart_station_idstart_station_namestart_latstart_lngend_station_idend_station_nameend_latend_lng...member_casualbike_share_for_all_triprental_access_methodrideable_typedateyearmonthday_of_weekis_weekendhour_of_day
02019-06-30 18:16:09.77302019-07-01 16:57:45.592010917th St at Valencia St37.763316-122.42190456Koshland Park37.773414-122.427317...memberNo<NA>NaN2019-06-30201966True18
12019-06-30 18:09:55.83002019-07-01 14:47:36.6810502nd St at Townsend St37.780526-122.39028810115th St at Potrero Ave37.767079-122.407359...casualNo<NA>NaN2019-06-30201966True18
22019-06-30 15:40:31.03802019-07-01 08:13:54.349023The Embarcadero at Steuart St37.791464-122.39103430San Francisco Caltrain (Townsend St at 4th St)37.776598-122.395282...memberNo<NA>NaN2019-06-30201966True15
42019-06-30 17:21:00.05502019-07-01 06:55:54.996015San Francisco Ferry Building (Harry Bridges Pl...37.795392-122.39420330San Francisco Caltrain (Townsend St at 4th St)37.776598-122.395282...casualNo<NA>NaN2019-06-30201966True17
62019-06-30 14:31:39.57302019-07-01 00:53:02.25206The Embarcadero at Sansome St37.804770-122.403234400Buchanan St at North Point St37.804272-122.433537...casualNo<NA>NaN2019-06-30201966True14
\n", "

5 rows × 21 columns

\n", "
" ], "text/plain": [ " started_at ended_at start_station_id \\\n", "0 2019-06-30 18:16:09.7730 2019-07-01 16:57:45.5920 109 \n", "1 2019-06-30 18:09:55.8300 2019-07-01 14:47:36.6810 50 \n", "2 2019-06-30 15:40:31.0380 2019-07-01 08:13:54.3490 23 \n", "4 2019-06-30 17:21:00.0550 2019-07-01 06:55:54.9960 15 \n", "6 2019-06-30 14:31:39.5730 2019-07-01 00:53:02.2520 6 \n", "\n", " start_station_name start_lat start_lng \\\n", "0 17th St at Valencia St 37.763316 -122.421904 \n", "1 2nd St at Townsend St 37.780526 -122.390288 \n", "2 The Embarcadero at Steuart St 37.791464 -122.391034 \n", "4 San Francisco Ferry Building (Harry Bridges Pl... 37.795392 -122.394203 \n", "6 The Embarcadero at Sansome St 37.804770 -122.403234 \n", "\n", " end_station_id end_station_name end_lat \\\n", "0 56 Koshland Park 37.773414 \n", "1 101 15th St at Potrero Ave 37.767079 \n", "2 30 San Francisco Caltrain (Townsend St at 4th St) 37.776598 \n", "4 30 San Francisco Caltrain (Townsend St at 4th St) 37.776598 \n", "6 400 Buchanan St at North Point St 37.804272 \n", "\n", " end_lng ... member_casual bike_share_for_all_trip \\\n", "0 -122.427317 ... member No \n", "1 -122.407359 ... casual No \n", "2 -122.395282 ... member No \n", "4 -122.395282 ... casual No \n", "6 -122.433537 ... casual No \n", "\n", " rental_access_method rideable_type date year month day_of_week \\\n", "0 NaN 2019-06-30 2019 6 6 \n", "1 NaN 2019-06-30 2019 6 6 \n", "2 NaN 2019-06-30 2019 6 6 \n", "4 NaN 2019-06-30 2019 6 6 \n", "6 NaN 2019-06-30 2019 6 6 \n", "\n", " is_weekend hour_of_day \n", "0 True 18 \n", "1 True 18 \n", "2 True 15 \n", "4 True 17 \n", "6 True 14 \n", "\n", "[5 rows x 21 columns]" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import datetime\n", "import seaborn as sns\n", "\n", "# convert timestamp into useful date/hour/day of week columns we can use\n", "\n", "def getDate(timestamp):\n", " date_only = timestamp.split()[0]\n", " return date_only\n", "\n", "def getYear(timestamp):\n", " date_only = timestamp.split()[0]\n", " date_arr = date_only.split('-')\n", " year = int(date_arr[0])\n", " return year\n", "\n", "def getMonth(timestamp):\n", " date_only = timestamp.split()[0]\n", " date_arr = date_only.split('-')\n", " month = int(date_arr[1])\n", " return month\n", "\n", "def getDayOfWeek(timestamp):\n", " date_only = timestamp.split()[0]\n", " date_arr = date_only.split('-')\n", " year = int(date_arr[0])\n", " month = int(date_arr[1])\n", " day = int(date_arr[2])\n", " return datetime.datetime(year, month, day).weekday()\n", "\n", "def getHourOfDay(timestamp):\n", " time_only = timestamp.split()[1]\n", " time_arr = time_only.split(':')\n", " return time_arr[0]\n", "\n", "def isWeekend(day_of_week):\n", " return day_of_week > 4\n", "\n", "lyft_data['date'] = lyft_data.apply(lambda row: getDate(row.started_at), axis=1)\n", "lyft_data['year']= lyft_data.apply(lambda row: getYear(row.started_at), axis=1)\n", "lyft_data['month'] = lyft_data.apply(lambda row: getMonth(row.started_at), axis=1)\n", "lyft_data['day_of_week'] = lyft_data.apply(lambda row: getDayOfWeek(row.started_at), axis=1)\n", "lyft_data['is_weekend'] = lyft_data.apply(lambda row: isWeekend(row.day_of_week), axis=1)\n", "lyft_data['hour_of_day'] = lyft_data.apply(lambda row: getHourOfDay(row.started_at), axis=1)\n", "\n", "lyft_data.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 1) Examining Temporal Factors\n", "\n", "Looking at usage over weekdays/weekends and over different times of day." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[Text(0, 0.5, ''),\n", " Text(0.5, 1.0, 'Figure 2: Distribution of Usage by Day of Week')]" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fig, axes = plt.subplots(2, 1, figsize=(12, 12), sharex=False)\n", "\n", "# group data based on weekday/weekend and hour of day\n", "hourly_usage = (lyft_data.groupby(['is_weekend'])['hour_of_day'].value_counts()\n", " .rename('usage count')\n", " .mul(100)\n", " .reset_index()\n", " .sort_values('hour_of_day'))\n", "\n", "hourly_usage = hourly_usage.replace({'is_weekend': True}, {'is_weekend': 'weekend'}, regex=True)\n", "hourly_usage = hourly_usage.replace({'is_weekend': False}, {'is_weekend': 'weekday'}, regex=True)\n", "\n", "\n", "hourly_usage_plot = sns.barplot(ax=axes[0], x=\"is_weekend\", y=\"usage count\", hue=\"hour_of_day\", data=hourly_usage)\n", "hourly_usage_plot.set(xlabel=None, title='Figure 1: Hourly Usage during Weekdays and Weekends')\n", "axes[0].legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.1, title='Hour of Day')\n", "\n", "# Plot usage data for each day of the week\n", "weekday_data = lyft_data.groupby(['day_of_week']).count()\n", "\n", "labels = ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday']\n", "weekday_plot = weekday_data.plot.pie(ax=axes[1], y='started_at', labels=labels, autopct='%1.1f%%')\n", "axes[1].get_legend().remove()\n", "weekday_plot.set(ylabel=None, title='Figure 2: Distribution of Usage by Day of Week')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The distribution of usage by day of week (*Figure 2*) shows that usage is slightly higher towards the weekend, but it's mostly equally distributed. However, the hourly usages vary quite a bit over the weekend versus weekday. \n", "\n", "According to *Figure 1*, weekday usage seems to peak when people get to work and when they get off work. Usage is also higher in the evenings, suggesting that people might use bike sharing as a more leisurely means of getting home, when they do not face the pressure of having to get to work on time (like in the mornings). On weekends, usage seems mirror people's leisure activities and peaks during the middle of the day. \n", "\n", "Next, let's compare the type of users over the weekend/weekday. " ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "weekday_usage_types = (lyft_data.groupby(['is_weekend'])['member_casual'].value_counts(normalize=True)\n", " .rename('percentage')\n", " .mul(100)\n", " .reset_index())\n", "\n", "weekday_usage_types = weekday_usage_types.replace({'is_weekend': True}, {'is_weekend': 'weekend'}, regex=True)\n", "weekday_usage_types = weekday_usage_types.replace({'is_weekend': False}, {'is_weekend': 'weekday'}, regex=True)\n", "\n", "sns.set(rc={'figure.figsize':(6, 6)})\n", "weekday_usage_types_plot = sns.barplot(x=\"is_weekend\", y=\"percentage\", hue=\"member_casual\", data=weekday_usage_types)\n", "weekday_usage_types_plot.set(xlabel=None, ylabel='Percentage (%)', title='Figure 3: Usage Types as a Percentage of Total Usage')\n", "plt.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0, title='Usage Type')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On weekdays, the share of casual versus subscribed (member) users are quite equally distributed. However, during weekends, casual users take up a much higher proportion of bike share usage. Based on the graphed usage by hour in *Figure 1*, usage seems to peak around evening commute hours, which suggests a siginificant amount of subscribed weekly commutters using bikes as a means to get home on a regular basis." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 2) Examining the role of weather\n", "\n", "Looking at each day's bike share usage, plotted against temperature and precipitation. I will look at monthly usage counts and monthly averages of temperature and precipitation. To examine this relationship further, I will plot average daily temperatures and precipitation against bike share usage counts. \n", "\n", "An outline of the steps below\n", "1. Get daily weather data (temperature and precipitation), convert to metric units\n", "2. Calculate the average monthly weather data, total up monthly bike share usage counts, and plot them by month.\n", "3. Group the Lyft bike share data into daily counts, plot these counts against daily temperature and precipitation" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
monthyeartemp_averageprecipitation
01201912.7688170.374445
11202011.5681000.200742
21202111.8189960.239252
3220199.9900790.652236
42202013.5344830.000000
\n", "
" ], "text/plain": [ " month year temp_average precipitation\n", "0 1 2019 12.768817 0.374445\n", "1 1 2020 11.568100 0.200742\n", "2 1 2021 11.818996 0.239252\n", "3 2 2019 9.990079 0.652236\n", "4 2 2020 13.534483 0.000000" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import calendar\n", "\n", "# read daily weather data files\n", "weather_files_by_year = ['sf_weather_data_2020.csv', 'sf_weather_data_2019.csv', 'sf_weather_data_2021.csv']\n", "li = []\n", "for filename in weather_files_by_year:\n", " df = pd.read_csv(filename, index_col=None, header=0)\n", " li.append(df)\n", " \n", "sf_weather = pd.concat(li, axis=0, ignore_index=True)\n", "sf_weather = sf_weather.drop(columns=['Maximum', 'Minimum', 'Departure', 'HDD', 'CDD'])\n", "sf_weather = sf_weather.rename(columns={'Date': 'date', 'Precipitation': 'precipitation', 'Average': 'temp_average'})\n", "\n", "def fahr_to_celsius(temp_fahr):\n", " temp_celsius = (temp_fahr - 32) * 5 / 9\n", " return temp_celsius\n", "\n", "def in_to_cm(precip_in):\n", " precip_cm = precip_in * 2.54\n", " return precip_cm\n", "\n", "def formatMonthYear(row):\n", " m = calendar.month_abbr[row['month']]\n", " y = str(row['year'])\n", " y = y[-2:]\n", " return m + \" '\" + y\n", "\n", "# convert units to metric (brrrr imperial)\n", "sf_weather['temp_average'] = sf_weather['temp_average'].astype(float)\n", "sf_weather['temp_average'] = fahr_to_celsius(sf_weather['temp_average'])\n", "sf_weather['precipitation'] = sf_weather['precipitation'].astype(float)\n", "sf_weather['precipitation'] = in_to_cm(sf_weather['precipitation'])\n", "\n", "\n", "sf_weather['month'] = pd.DatetimeIndex(sf_weather['date']).month\n", "sf_weather['year'] = pd.DatetimeIndex(sf_weather['date']).year\n", "sf_weather['month_year'] = sf_weather.apply(lambda row: formatMonthYear(row), axis=1)\n", "\n", "# get each month's average temperature (sf_weather is our daily average)\n", "sf_weather_monthly_average = (sf_weather.groupby(['month', 'year']).mean()\n", " .reset_index())\n", "\n", "sf_weather_monthly_average.head()" ] }, { "cell_type": "code", "execution_count": 242, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[Text(0, 0.5, 'usage count'),\n", " Text(0.5, 0, 'precipitation (cm) / temp (c)'),\n", " Text(0.5, 1.0, 'Figure 5: Weather and Daily Usage')]" ] }, "execution_count": 242, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, axes = plt.subplots(2, 1, figsize=(20, 15))\n", "\n", "\n", "\n", "###################### plot monthly bike share data with weather ###################### \n", "\n", "# get monthly bike share totals\n", "usage_by_month = (lyft_data.groupby(['month', 'year']).size()\n", " .rename('usage count')\n", " .reset_index())\n", "usage_by_month['month_year'] = usage_by_month.apply(lambda row: formatMonthYear(row), axis=1)\n", "\n", "# sort bars by year, then month\n", "usage_by_month = usage_by_month.sort_values(['year', 'month']).reset_index().drop(columns=['index'])\n", "usage_by_month = usage_by_month.reset_index().rename(columns={'index': 'order'})\n", "usage_by_month['month_year'] = pd.Categorical(usage_by_month['month_year'],\n", " categories=usage_by_month['month_year'].tolist(),\n", " ordered=True)\n", "# merge monthly usage data and monthly weather averages\n", "monthly_usage_and_weather = pd.merge(usage_by_month, sf_weather_monthly_average)\n", "\n", "# plot monthly usage, temperature, and precipitation data on the same plot\n", "monthly_usage_plot = sns.barplot(ax=axes[0], data=monthly_usage_and_weather, x='month_year', y='usage count', color='b')\n", "ax2 = axes[0].twinx()\n", "monthly_weather_plot = sns.lineplot(data=monthly_usage_and_weather, x='month_year', y='temp_average', color=\"r\", ax=ax2)\n", "axes[0].grid(False)\n", "ax2.grid(False)\n", "monthly_weather_plot.set(xlabel='month', ylabel='precipitation (cm) / temp (c)', title='Figure 4: Monthly Usage and Weather Averages')\n", "\n", "\n", "\n", "###################### showing the relationship between daily bike share counts with weather ###################### \n", "\n", "# get daily bike share totals\n", "lyft_data_daily = (lyft_data.groupby(['date']).size()\n", " .rename('count')\n", " .reset_index())\n", "\n", "# merge daily usage data with daily weather averages\n", "lyft_data_daily = pd.merge(lyft_data_daily, sf_weather)\n", "lyft_data_daily\n", "\n", "daily_weather_and_usage_plot = sns.scatterplot(ax=axes[1], data=lyft_data_daily, x=\"temp_average\", y=\"count\")\n", "sns.scatterplot(ax=axes[1], data=lyft_data_daily, x=\"precipitation\", y=\"count\")\n", "daily_weather_and_usage_plot.set(xlabel='precipitation (cm) / temp (c)', ylabel='usage count', title='Figure 5: Weather and Daily Usage')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Based on Figure 4, usage seems to decrease with extremely low temperatures. From October to December of 2020 and 2021, usage and temperatures both fall together. There is a notable spike in usage in January and February of 2020, which can be attributed to the [return of electric bikes](https://www.sfexaminer.com/news/lyft-set-to-return-e-bike-to-sf-streets-after-solving-battery-fire-issue/) after battery issues in July 2019 caused Lyft to pause their electric fleet offerings. From March 19, 2020 to January 25, 2021, California issued a statewide stay-at-home order, which could explain a drop in the usage. Besides the spike from the return of electric bikes at the beginning of 2020, there doesn't seem to be a lot of flucuation in usage in the monthly data. \n", "\n", "Figure 5 is more telling of the relationship between weather and usage. Although San Francisco does not get a lot of rain at all, users are still very sensitive to precipitation levels. At low levels of precipitation, usage is under 10,000 rides per day. At slightly higher levels of precipitation (around 2 – 4cm per day), usage does not go avove 6000 rides per day. Temperatures within the modereate, regular range (of around 10 – 18 °C) does not seem to have much of an effect on bike share usage. However, when temperatures gets warmer (above 20 °C), usage drops to below 8000 rides per day.\n", "\n", "These results confirm Eren and Uz (2020)’s findings to an extent. These results confirm Eren and Uz's findings that trip production is positively correlated with temperature when temperature ranges around 0 – 20 °C with no precipitation. Eren and Uz also found that 20 – 30 °C is the temperature range in which temperature bike sharing demand is at the maximum level. These results show that bike sharing demand fluctuates to higher levels around 10 - 20 °C, but the positive correlation is still observed past 20 °C." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 3) Examing Safety and Public Transportation as Factors Affeting Bike Sharing\n", "\n", "Mapping crime rate, public transportation options, and bike share usage" ] }, { "cell_type": "code", "execution_count": 241, "metadata": { "scrolled": false }, "outputs": [ { "data": { "text/html": [ "
Make this Notebook Trusted to load map: File -> Trust Notebook
" ], "text/plain": [ "" ] }, "execution_count": 241, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import shapely\n", "from folium.plugins import BeautifyIcon\n", "import folium\n", "\n", "sf_latitude = 37.7749\n", "sf_longitude = -122.4194\n", "\n", "# Create map\n", "sf_usage_map = folium.Map(\n", " location=[sf_latitude,sf_longitude],\n", " zoom_start=13,\n", " tiles='cartodb positron')\n", "\n", "###################### map out crime rate per neighborhood ###################### \n", "\n", "# read sf neighborhood polygons\n", "neighborhoods = gpd.read_file('SF_Analysis_Neighborhoods.csv')\n", "neighborhoods = neighborhoods.drop(columns=['geometry'])\n", "neighborhoods = neighborhoods.rename(columns={'NHOOD': 'Analysis Neighborhood', 'the_geom': 'geometry'})\n", "\n", "# read sf crime data\n", "incidents = pd.read_csv('SF_Police_Incident_Reports_2018_to_Present.csv') \n", "incidents = incidents.dropna(subset=['Latitude', 'Longitude'])\n", "\n", "# get a count of incidents by neighborhood, sorted from lowest to highest\n", "incidents_by_neighborhood = (incidents.groupby(['Analysis Neighborhood']).size()\n", " .rename('count')\n", " .reset_index()\n", " .sort_values('count'))\n", "\n", "# add color based on number of incidents (for map) \n", "incidents_by_neighborhood['fill_color'] = pd.cut(incidents_by_neighborhood['count'], bins=5, \n", " labels=['#fcfadc', '#F2E750', '#F2B807', '#F28705', '#C52104'])\n", "\n", "\n", "# get each neighborhood's polygons from the neighborhood dataset\n", "incidents_by_neighborhood = pd.merge(incidents_by_neighborhood, neighborhoods)\n", "\n", "# convert polygon data for mapping\n", "geometry = incidents_by_neighborhood['geometry'].map(shapely.wkt.loads)\n", "incidents_by_neighborhood = incidents_by_neighborhood.drop('geometry', axis=1)\n", "neighborhood_incidents_gdf = gpd.GeoDataFrame(incidents_by_neighborhood, crs=\"EPSG:4326\", geometry=geometry)\n", "\n", "\n", "folium.GeoJson(neighborhood_incidents_gdf,\n", " style_function=lambda feature: {\n", " 'fillColor': feature['properties']['fill_color'],\n", " 'color' : feature['properties']['fill_color'],\n", " 'fillOpacity' : 0.3,\n", " }).add_to(sf_usage_map)\n", "\n", "\n", "###################### add public transit stations data ###################### \n", "\n", "stations = pd.read_csv('Passenger_Rail_Stations_2019.csv')\n", "\n", "# add MUNI stations (light rail)\n", "muni_stations = stations[stations['mode'] == 'Light Rail']\n", "for index, row in muni_stations.iterrows():\n", " lng = row.X\n", " lat = row.Y\n", " folium.CircleMarker(\n", " [lat,lng],\n", " radius=1,\n", " color='#d1feff',\n", " fill=False,\n", " fill_opacity=1\n", " ).add_to(sf_usage_map) \n", "\n", "# add Cal Train stations\n", "cal_train_stations = stations[stations['mode'] == 'Commuter Rail']\n", "for index, row in cal_train_stations.iterrows():\n", " lng = row.X\n", " lat = row.Y\n", " folium.CircleMarker(\n", " [lat,lng],\n", " radius=8,\n", " color='#9768D1',\n", " fill=True,\n", " fill_color='#9768D1',\n", " fill_opacity=1\n", " ).add_to(sf_usage_map) \n", "\n", "# add BART stations (subway)\n", "bart_stations = stations[stations['mode'] == 'Rapid Rail']\n", "for index, row in bart_stations.iterrows():\n", " lng = row.X\n", " lat = row.Y\n", " folium.CircleMarker(\n", " [lat,lng],\n", " radius=5,\n", " color='#9C27B0',\n", " fill=True,\n", " fill_color='#9C27B0',\n", " fill_opacity=1\n", " ).add_to(sf_usage_map) \n", "\n", "###################### add bike usage data ###################### \n", "\n", "# get total usage by station, color marker based on usage\n", "loc_data = lyft_data.groupby(['start_station_id']).size().rename('count').reset_index().sort_values('count')\n", "\n", "marker_colors = ['#b1ff91', '#94e586', '#7acb7b', '#4b9963', '#27b48f',\n", " '#378056', '#246848', '#14513a', '#063b2b', '00261c']\n", "loc_data['marker_color'] = pd.cut(loc_data['count'], bins=10, \n", " labels=marker_colors)\n", "\n", "\n", "# getlat lon of each station \n", "lyft_data = lyft_data.dropna(subset=['start_station_id'])\n", "loc_data_first_row = lyft_data.groupby(['start_station_id']).head(1)\n", "loc_data_first_row = loc_data_first_row.drop(\n", " columns=['started_at', 'ended_at', 'bike_id', 'hour_of_day', 'day_of_week', 'is_weekend', 'rideable_type', 'member_casual'])\n", "\n", "loc_data = pd.merge(loc_data, loc_data_first_row ,on='start_station_id',how='inner')\n", "\n", "for lat,lng, marker_color in zip(loc_data.start_lat, loc_data.start_lng, loc_data.marker_color):\n", " folium.CircleMarker(\n", " [lat,lng],\n", " radius=3,\n", " color=marker_color,\n", " fill=True,\n", " fill_color=marker_color,\n", " fill_opacity=0.5\n", " ).add_to(sf_usage_map) \n", "\n", "\n", "sf_usage_map" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Map Legend:\n", "\n", "MUNI Station         \n", "\n", "Cal Train Station         \n", "\n", "\n", "BART Station         \n", "\n", "Usage at each station, from LEAST to GREATEST: \n", "       \n", "       \n", "       \n", "       \n", "       \n", "       \n", "       \n", "       \n", "       \n", "       \n", "       \n", "\n", "Crime rate per neighborhood, from LOWEST to HIGHEST:\n", "       \n", "       \n", "       \n", "       \n", "       \n", "       \n", "\n", "#### Analysis\n", "\n", "From the map, it looks like bike shares are being used most heavily around BART (the rail/subway system) and Cal Train stations as a complement to the users' means of getting home. However, the same usage pattern does not seem to hold for the MUNI light rail. This could be due to the fact that MUNI stops are close together and that MUNI trips do not go as far, which means that MUNI serves the same purpose as the bikes. Therefore, the bikes act more like a substitute to MUNI than a complement. There is also a fair amoutn usage of bike shares around areas with only bus services (the light yellow area at the center of the map. \n", "\n", "These findings are in line with Eren and Uz's (2020) fidings that bike shares can act as both a complement to public transportation to reduce travel time and a substitute in areas with poor public transport options. \n", "\n", "\n", "There seems to be less of a concern for safety (measured by crime rates) factored into the decision to use bike share services. Areas with the highest crime rates are also the busiest transportation hubs, so usage is necessary in many cases. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 4) Predict \n", "\n", "Given weekday/weekend, temperature, precipitation, hour of day, bike type (electric scooter, docked bike, classic bike), predict the demand for bike shares. " ] }, { "cell_type": "code", "execution_count": 143, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
is_weekendcounttemp_averageprecipitationmonthmember_casual_memberhour_of_day_01hour_of_day_02hour_of_day_03hour_of_day_04...hour_of_day_16hour_of_day_17hour_of_day_18hour_of_day_19hour_of_day_20hour_of_day_21hour_of_day_22hour_of_day_23rideable_type_docked_bikerideable_type_electric_bike
00.03.011.3888890.04.000000...0000000001
10.06.011.3888890.04.001000...0000000001
\n", "

2 rows × 31 columns

\n", "
" ], "text/plain": [ " is_weekend count temp_average precipitation month \\\n", "0 0.0 3.0 11.388889 0.0 4.0 \n", "1 0.0 6.0 11.388889 0.0 4.0 \n", "\n", " member_casual_member hour_of_day_01 hour_of_day_02 hour_of_day_03 \\\n", "0 0 0 0 0 \n", "1 0 1 0 0 \n", "\n", " hour_of_day_04 ... hour_of_day_16 hour_of_day_17 hour_of_day_18 \\\n", "0 0 ... 0 0 0 \n", "1 0 ... 0 0 0 \n", "\n", " hour_of_day_19 hour_of_day_20 hour_of_day_21 hour_of_day_22 \\\n", "0 0 0 0 0 \n", "1 0 0 0 0 \n", "\n", " hour_of_day_23 rideable_type_docked_bike rideable_type_electric_bike \n", "0 0 0 1 \n", "1 0 0 1 \n", "\n", "[2 rows x 31 columns]" ] }, "execution_count": 143, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# prep/clean up data for prediction:\n", "\n", "lyft_data['is_bart'] = lyft_data['start_station_name'].map(lambda station_name: True if 'BART' in str(station_name) else False)\n", "\n", "daily_totals = (lyft_data.groupby(['date', 'member_casual', 'is_weekend', 'hour_of_day', 'rideable_type']).size()\n", " .rename('count')\n", " .reset_index())\n", "\n", "daily_totals = pd.merge(daily_totals, sf_weather)\n", "daily_totals = daily_totals.drop(columns=['month_year', 'year', 'date'])\n", "\n", "daily_totals.loc['hour_of_day'] = daily_totals['hour_of_day'].astype(str)\n", "daily_totals = pd.get_dummies(data=daily_totals, drop_first=True)\n", "daily_totals = daily_totals.dropna()\n", "\n", "daily_totals.head(2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Run Linear, Ridge, and Lasso Regressions and compare results. Run 1000 iterations of each to get an average score. " ] }, { "cell_type": "code", "execution_count": 276, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Linear Regression score: 0.5725204848937282\n", "Mean Absolute Error: 19.9304171877794\n", "Mean Squared Error: 890.5226432325926\n", "Root Mean Squared Error: 29.841626015225653\n", "\n", "Ridge Regression 0.5732229188055942\n", "Mean Absolute Error: 19.712357309825105\n", "Mean Squared Error: 845.6348161960889\n", "Root Mean Squared Error: 29.07980082799896\n", "\n", "Lasso Regression 0.572997036210082\n", "Mean Absolute Error: 19.45921781445111\n", "Mean Squared Error: 826.0696322627631\n", "Root Mean Squared Error: 28.74142710901397\n" ] } ], "source": [ "from sklearn.model_selection import train_test_split\n", "from sklearn import linear_model\n", "from sklearn.linear_model import LinearRegression\n", "from sklearn import metrics\n", "\n", "X = np.asarray(daily_totals.drop(columns=['count']))\n", "Y = np.asarray(daily_totals['count'])\n", "\n", "scores = []\n", "coefs = []\n", "for i in range(1000):\n", " X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.33, shuffle= True)\n", " lineReg = LinearRegression()\n", " lineReg.fit(X_train, y_train)\n", " scores.append(lineReg.score(X_test, y_test))\n", " coefs.append(lineReg.coef_)\n", "print('Linear Regression score:', np.mean(scores))\n", "y_pred = lineReg.predict(X_test)\n", "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))\n", "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))\n", "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))\n", "\n", "scores = []\n", "coefs = []\n", "for i in range(1000):\n", " X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.33, shuffle= True)\n", " lineRegRidge = linear_model.Ridge (alpha = .5)\n", " lineRegRidge.fit(X_train, y_train)\n", " scores.append(lineReg.score(X_test, y_test))\n", " coefs.append(lineReg.coef_)\n", "y_pred = lineReg.predict(X_test) \n", "print('\\nRidge Regression', np.mean(scores))\n", "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))\n", "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))\n", "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))\n", "\n", "scores = []\n", "coefs = []\n", "for i in range(1000):\n", " X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.33, shuffle= True)\n", " lineRegLasso = linear_model.Lasso (alpha = .1)\n", " lineRegLasso.fit(X_train, y_train)\n", " scores.append(lineReg.score(X_test, y_test))\n", " coefs.append(lineReg.coef_)\n", "y_pred = lineReg.predict(X_test)\n", "print('\\nLasso Regression', np.mean(scores))\n", "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))\n", "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))\n", "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Linear, Ridge, and Lasso Regression scores are quite similar, but none are looking that good yet. Let's see a plot.**" ] }, { "cell_type": "code", "execution_count": 274, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 274, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "sns.regplot(y_test, lineReg.predict(X_test))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Let's try applying a polynomial transformation to X.**" ] }, { "cell_type": "code", "execution_count": 277, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "Polymial Transformation + Linear Regression: 0.8001068259671344\n", "Mean Absolute Error: 13.116596688775768\n", "Mean Squared Error: 417.12276607927186\n", "Root Mean Squared Error: 20.423583575838787\n" ] } ], "source": [ "from sklearn.preprocessing import PolynomialFeatures\n", "\n", "# polynomial transformation (with degree 3)\n", "\n", "poly = PolynomialFeatures(degree=2)\n", "poly_X = poly.fit_transform(X)\n", "\n", "scores = []\n", "coefs = []\n", "for i in range(1):\n", " poly_X_train, poly_X_test, y_train, y_test = train_test_split(poly_X, Y, test_size = 0.3, shuffle= True, random_state = 4)\n", " lineReg = LinearRegression()\n", " lineReg.fit(poly_X_train, y_train)\n", " scores.append(lineReg.score(poly_X_test, y_test))\n", " coefs.append(lineReg.coef_)\n", "\n", "print('\\nPolymial Transformation + Linear Regression score:', np.mean(scores))\n", "y_pred = lineReg.predict(poly_X_test)\n", "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))\n", "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))\n", "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is looking better now. Lets see our plot and coefficients.\n", "\n", "**Coefficients of our polynomial features:**" ] }, { "cell_type": "code", "execution_count": 278, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
1is_weekendtemp_averageprecipitationmonthmember_casual_memberhour_of_day_01hour_of_day_02hour_of_day_03hour_of_day_04...hour_of_day_22^2hour_of_day_22 hour_of_day_23hour_of_day_22 rideable_type_docked_bikehour_of_day_22 rideable_type_electric_bikehour_of_day_23^2hour_of_day_23 rideable_type_docked_bikehour_of_day_23 rideable_type_electric_bikerideable_type_docked_bike^2rideable_type_docked_bike rideable_type_electric_bikerideable_type_electric_bike^2
01.00.011.3888890.04.00.00.00.00.00.0...0.00.00.00.00.00.00.00.00.01.0
11.00.011.3888890.04.00.01.00.00.00.0...0.00.00.00.00.00.00.00.00.01.0
21.00.011.3888890.04.00.00.01.00.00.0...0.00.00.00.00.00.00.01.00.00.0
31.00.011.3888890.04.00.00.01.00.00.0...0.00.00.00.00.00.00.00.00.01.0
41.00.011.3888890.04.00.00.00.01.00.0...0.00.00.00.00.00.00.00.00.01.0
\n", "

5 rows × 496 columns

\n", "
" ], "text/plain": [ " 1 is_weekend temp_average precipitation month member_casual_member \\\n", "0 1.0 0.0 11.388889 0.0 4.0 0.0 \n", "1 1.0 0.0 11.388889 0.0 4.0 0.0 \n", "2 1.0 0.0 11.388889 0.0 4.0 0.0 \n", "3 1.0 0.0 11.388889 0.0 4.0 0.0 \n", "4 1.0 0.0 11.388889 0.0 4.0 0.0 \n", "\n", " hour_of_day_01 hour_of_day_02 hour_of_day_03 hour_of_day_04 ... \\\n", "0 0.0 0.0 0.0 0.0 ... \n", "1 1.0 0.0 0.0 0.0 ... \n", "2 0.0 1.0 0.0 0.0 ... \n", "3 0.0 1.0 0.0 0.0 ... \n", "4 0.0 0.0 1.0 0.0 ... \n", "\n", " hour_of_day_22^2 hour_of_day_22 hour_of_day_23 \\\n", "0 0.0 0.0 \n", "1 0.0 0.0 \n", "2 0.0 0.0 \n", "3 0.0 0.0 \n", "4 0.0 0.0 \n", "\n", " hour_of_day_22 rideable_type_docked_bike \\\n", "0 0.0 \n", "1 0.0 \n", "2 0.0 \n", "3 0.0 \n", "4 0.0 \n", "\n", " hour_of_day_22 rideable_type_electric_bike hour_of_day_23^2 \\\n", "0 0.0 0.0 \n", "1 0.0 0.0 \n", "2 0.0 0.0 \n", "3 0.0 0.0 \n", "4 0.0 0.0 \n", "\n", " hour_of_day_23 rideable_type_docked_bike \\\n", "0 0.0 \n", "1 0.0 \n", "2 0.0 \n", "3 0.0 \n", "4 0.0 \n", "\n", " hour_of_day_23 rideable_type_electric_bike rideable_type_docked_bike^2 \\\n", "0 0.0 0.0 \n", "1 0.0 0.0 \n", "2 0.0 1.0 \n", "3 0.0 0.0 \n", "4 0.0 0.0 \n", "\n", " rideable_type_docked_bike rideable_type_electric_bike \\\n", "0 0.0 \n", "1 0.0 \n", "2 0.0 \n", "3 0.0 \n", "4 0.0 \n", "\n", " rideable_type_electric_bike^2 \n", "0 1.0 \n", "1 1.0 \n", "2 0.0 \n", "3 1.0 \n", "4 1.0 \n", "\n", "[5 rows x 496 columns]" ] }, "execution_count": 278, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "sns.regplot(y_test, lineReg.predict(poly_X_test))\n", "\n", "data = daily_totals.drop(columns=['count'])\n", "cols = data.columns.values\n", "features = pd.DataFrame(poly.transform(data), columns=poly.get_feature_names(cols))\n", "features.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Conclusion\n", "\n", "Of the factors examined, timing (time of day, weekday/weekend) and public transportation are the biggest factors affecting bike sharing in San Francisco. \n", "\n", "The findings that temperatures from the 10-25 range with no precipitation yields a positive correlation with bike share demand are in line with Eren and Uz's (2020) findings. It also confirms the hypothesis that San Francisco's change in temperature and precipitation over each year is mild enough that seasonality (time of year) does not play a role in bike sharing demand. \n", "\n", "These findings also support the hypothesis and Eren and Uz's (2020) findings that bike shares can act as both a complement to public transportation to reduce travel time and a substitute in areas with poor public transport options.\n", "\n", "However, these findings do not support the hypothesis about the negative correlation between bike share usage and crime rates. Crime rate seems to have no effect on bike sharing usage in San Francisco. \n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Limitations\n", "\n", "##### COVID-19\n", "Bike share usage has been affected by the Caifornia's stay-at-home orders and changes to movement patterns as people started working from home from March, 2020. \n", "\n", "##### Lyft's eBike Battery Issues\n", "Lyft's battery issues caused Lyft to temporarily suspend their offering of electric bikes from July to mid-December of 2019. As electric bikes are their most popular offering, this caused a drop in usage during the period which these eBikes we not avaialble. When they were put back in January 2020, usage seemed to skyrocket until California announced its statewide stay-at-home order. \n", "\n", "##### Demographics\n", "Many studies examined income and demographics (e.g. age, gender) as significant factors that affect bike share usage. Lyft's anonymized data contains no information on demographics." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Areas for further research\n", "\n", "##### Usage Patterns\n", "It would be helpful to further examine usage patterns (duration, distance, start/end locations) to see *how* people are using bike shares (e.g. Where do those who use bike sharing as a means to get home from a BART station live? Are there any other public transportation options avialable for them?). Is it for convenience, cost, health, leisure or other factors?\n", "\n", "##### Electric versus Regular Bikes\n", "The usage of electric bikes are much higher in San Francisco, possibly due to the fact that the city has quite a lot of steeper hills. I would examine the usage patterns of electric versus regular bikes with the user's path to see whether elevation gain is a factor in choosing an electric bike. I would also like to examine whether there is a higher usage of electric bikes when bike sharing is used as a means to get around rather than for leisure (e.g. riding through an area with high crime rates, weekdays, rush hours, late night, riding from a transit stop). " ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.7.6" } }, "nbformat": 4, "nbformat_minor": 4 }