Bike sharing
5992e388d3cb9 bpfull Bike sharing systems are a means of renting bicycles where the process of obtaining membership, rent...
2018
16/03
 
  Partecipanti 53 Sottomissioni 969  
 

Bike sharing systems are a means of renting bicycles where the process of obtaining membership, rental, and bike return is automated via a network of kiosk locations throughout a city. Using these systems, people are able rent a bike from a one location and return it to a different place on an as-needed basis. Currently, there are over 500 bike-sharing programs around the world.

The data generated by these systems makes them attractive for researchers because the duration of travel, departure location, arrival location, and time elapsed is explicitly recorded. Bike sharing systems therefore function as a sensor network, which can be used for studying mobility in a city. In this competition, participants are asked to combine historical usage patterns with weather data in order to forecast bike rental demand.

Submissions are evaluated by the Root Mean Squared Logarithmic Error (RMSLE). The RMSLE is calculated as

RMSE sqrt( mean( (log(y+1) – log(yhat+1) )^2 ) )

Where:

yhat is your predicted count
y is the actual count
log(x) is the natural logarithm

During the competition, the leaderboard displays your partial score, which is the RMSE for 3000 (random) observations of the test set.
At the end of the contest, the leaderboard will display the final score, which is the RMSE for the remaining 3493 observations of the test set. The final score will determine the final winner. This method prevents users from overfitting to the leaderboard.

Team max size = 3

library(fasttime)
library(lubridate)
library(rpart)

train <- read.csv(“99.csv”, stringsAsFactors=T)
test <- read.csv(“100.csv”, , stringsAsFactors=F)
train$count = log1p(train$count)

train >
mutate(datetime = fastPOSIXct(datetime, “GMT”)) >
mutate(hour = hour(datetime),
month = month(datetime),
year = year(datetime),
wday = wday(datetime)) → train

test >
mutate(datetime = fastPOSIXct(datetime, “GMT”)) >
mutate(hour = hour(datetime),
month = month(datetime),
year = year(datetime),
wday = wday(datetime)) → test

fml <- count ~ season + holiday + workingday + weather + temp + atemp + humidity + hour
fit <- rpart(fml, data=train)
yhat <- expm1(predict(fit, newdata=test))
write.table(file=“mybike.txt”, yhat, row.names = FALSE, col.names = FALSE)

datetime – hourly date + timestamp
season – 1 = spring, 2 = summer, 3 = fall, 4 = winter
holiday – whether the day is considered a holiday
workingday – whether the day is neither a weekend nor holiday
weather –
1: Clear, Few clouds, Partly cloudy, Partly cloudy
2: Mist + Cloudy, Mist + Broken clouds, Mist + Few clouds, Mist
3: Light Snow, Light Rain + Thunderstorm + Scattered clouds, Light Rain + Scattered clouds
4: Heavy Rain + Ice Pallets + Thunderstorm + Mist, Snow + Fog
temp – temperature in Celsius
atemp – “feels like” temperature in Celsius
humidity – relative humidity
windspeed – wind speed
casual – number of non-registered user rentals initiated
registered – number of registered user rentals initiated
count – number of total rentals (RESPONSE VARIABLE)




train_bike train_bike.csv
700 KB
test_bike test_bike.csv
300 KB
Per partecipare bisogna prima autenticarsi
# Nome Punteggio Prove Ultima prova
1 chiara.aldeghi95 FINALE 38.08% 27 15.11.2018
16:00
2 i.iacoban FINALE 38.08% 15 15.11.2018
16:10
3 c.crippa19 FINALE 38.08% 11 15.11.2018
16:12
4 Elisa Pirotta FINALE 38.18% 35 16.11.2018
09:41
5 NicholasMissineo FINALE 38.18% 32 16.11.2018
07:17
6 beatrice.somaschini21 FINALE 38.18% 30 16.11.2018
05:47
7 fabio.marigo FINALE 38.24% 32 14.11.2018
17:06
8 l.mandelli17 FINALE 38.24% 20 14.11.2018
17:32
9 giabellianna FINALE 38.24% 14 15.11.2018
09:33
10 e.bagnati FINALE 38.48% 29 15.11.2018
07:49
11 sarasixti FINALE 38.48% 4 15.11.2018
11:53
12 m.caronte FINALE 38.48% 1 15.11.2018
08:33
13 r.buzzini FINALE 38.62% 24 15.11.2018
13:40
14 g.floriani FINALE 38.62% 5 15.11.2018
14:36
15 d.parimbelli2 FINALE 39.00% 65 15.11.2018
09:21
16 d.casiraghi FINALE 39.00% 7 15.11.2018
08:57
17 a.gaffuririva FINALE 39.08% 14 15.11.2018
22:31
18 giorgia.modafferi FINALE 39.08% 1 15.11.2018
23:24
19 marcello.sbordi FINALE 39.08% 1 16.11.2018
08:16
20 s.turi FINALE 39.12% 32 13.11.2018
20:14
21 g.grazzi FINALE 39.12% 9 15.11.2018
07:41
22 d.casamassima FINALE 39.32% 39 15.11.2018
22:52
23 m.abbiati FINALE 39.32% 13 16.11.2018
08:23
24 d.perrini FINALE 39.39% 27 16.11.2018
03:18
25 m.sciartilli FINALE 39.39% 19 11.11.2018
23:08
26 s.panizza5 FINALE 39.46% 28 16.11.2018
08:05
27 f.nigro5 FINALE 39.46% 16 16.11.2018
08:08
28 f.bekollari FINALE 39.52% 8 13.11.2018
13:14
29 a.giampino FINALE 39.52% 2 03.11.2018
07:20
30 a.ongaro3 FINALE 39.52% 2 14.11.2018
13:45
31 l.bassanese FINALE 39.60% 24 16.11.2018
09:47
32 f.melograna1 FINALE 40.14% 47 16.11.2018
07:49
33 k.arguirov FINALE 40.14% 23 15.11.2018
22:17
34 e.benincasa1 FINALE 40.14% 3 15.11.2018
22:04
35 i.bessone FINALE 41.02% 13 15.11.2018
13:03
36 l.gregori1 FINALE 41.02% 5 15.11.2018
15:39
37 alessandra.pellegata FINALE 41.02% 1 15.11.2018
12:49
38 a.spataro2 FINALE 41.25% 35 15.11.2018
23:35
39 i.belleri FINALE 41.25% 10 16.11.2018
07:28
40 e.pelagalli FINALE 41.25% 49 15.11.2018
21:08
41 d.piovesana FINALE 41.25% 16 16.11.2018
09:46
42 Francesco Bongini FINALE 43.24% 7 30.10.2018
15:27
43 a.caciolo FINALE 43.62% 19 15.11.2018
17:43
44 n.ghioldi FINALE 43.62% 2 15.11.2018
15:45
45 p.dangelo4 FINALE 43.90% 6 14.11.2018
14:41
46 valyde FINALE 45.38% 6 16.11.2018
08:02
47 l.giordano8 FINALE 45.38% 4 16.11.2018
08:18
48 s.pisaniello2 FINALE 46.00% 32 16.11.2018
10:30
49 sonia_cucchi FINALE 46.00% 10 16.11.2018
09:46
50 f.logiudice1 FINALE 46.65% 5 15.11.2018
17:45
51 g.saccaggi FINALE 51.59% 63 15.11.2018
17:45
52 joana.curri FINALE 51.59% 22 16.11.2018
08:25
53 benchmark FINALE 72.50% 5 13.11.2018
17:18