Description
The bike-sharing rental process is highly correlated to the environmental and seasonal settings. For instance,
weather conditions, precipitation, day of week, season, hour of the day, etc. can affect the rental behaviors. The
core data set for this project is related to the two-year historical log corresponding to years 2011 and 2012 from
Capital Bikeshare System, Washington D.C. The data contains conditions (11 xi’s) on a daily basis along with the
count of bikes that were rented each day (y).
season : season (1:springer, 2:summer, 3:fall, 4:winter)
yr : year (0: 2011, 1:2012)
mnth : month ( 1 to 12)
holiday : whether a day is a holiday or not (0=no, 1 = yes)
weekday : day of the week (0-6)
workingday : if day is neither a weekend nor holiday, = 1, otherwise = 0.
weathersit :
o 1: Clear, Few clouds, Partly cloudy, Partly cloudy
o 2: Mist + Cloudy, Mist + Broken clouds, Mist + Few clouds, Mist
o 3: Light Snow, Light Rain + Thunderstorm + Scattered clouds, Light Rain + Scattered clouds
o 4: Heavy Rain + Ice Pellets + Thunderstorm + Mist, Snow + Fog
temp : Normalized temperature in Celsius. The values are divided to 41 (max)
atemp: Normalized feeling temperature in Celsius. The values are divided to 50 (max)
hum: Normalized humidity. The values are divided to 100 (max)
windspeed: Normalized wind speed. The values are divided to 67 (max)
cnt: count of total rental bikes including both casual and registered
Your assignment is to develop a regression model that will predict how many bikes will be rented on any
particular day based on these 11 conditions. Your data set is named BikeAll.txt and is formatted as follows:
first line: integer number of lines of training data (m = 731), tab, integer number of features (n = 11)
all other lines: values of input features, tab separated, followed by the number of bikes rented that day
(in order listed above)
Example:
m n
x
(1)
1 x
(1)
2 x
(1)
3 . . . x(1)
n y
(1)
. . .
x
(m)
1 x
(m)
2 x
(m)
3 . . . x(m)
n y
(m)
You should develop a model and hypothesis function that provide the best predictions of how many bikes will be
rented. In order for us to evaluate your model it must follow strict input and output requirements.
Your Python program must run in Spyder. It should first prompt the use for a Training Set file name that is
formatted as:
first line: integer number of lines of training data (m), tab, integer number of features (n)
all other lines: values of input features, tab separated, followed by the number of bikes rented that day
(in order listed above)
Next: Prints out values for all weights and final J for the training set. All output should be clearly
labelled.
Next: Prompts the user for the name of a validation set. Same file format as the training data.
Next: Prints out J for the validation set. All clearly labelled.
Next: Prompts the user for the name of a test set. Same file format as the training data.
Next: Prints out J for the test set and adjusted R2
for the test set. All clearly labelled.
Model development.
You should randomize the data file and put 439 data sets in your training set file, 146 data sets in your cross
validation set data file and 146 data sets in your test set data file.
You can choose to use the training data features as given or to create additional features. For example, if the
original feature data was:
x1, x2, x3
you could square each value to create a new training set consisting of
x1, x1
2
, x2, x2
2
, x3, x3
2
The important thing to remember is that the different model should be represented entirely by your data file!
Your program should be constructed so that it will work for any training set that lists, m, n on the first line and
data sets on each following line.
What to turn in (uploaded to canvas).
1. A pdf file containing:
A table describing your different models and giving your J values for test and validation sets for
all models, and J and R2
values for the test set for the best model.
A list of your final weights.
Example of table:
Model Jtrain Jvalidation Jtest/R2
test
Linear (n=11) 440,000 460,000
Quadratic
(n=22)
290,000 295,000 296,000/0.84
Other models
2. Your Python program named: yourlastname_yourfirstname_P1.py
3. Your Final Training set named: yourlastname_yourfirstname_Train.txt
4. Your Final Validation set named: yourlastname_yourfirstname_Valid.txt
5. Your Final Test set named: yourlastname_yourfirstname_Test.txt
Grading
Pdf file with all information included +20 +20
All py and text files turned in with all
information and correctly named
+20 +20
Validation set output CPSC 4820 CPSC 6820
J < 450,000 +55 +45
J < 300,000 +60 +55
J < 250,000 +65 +65
Program throws an error -65 -65
Program does not prompt for training set -10 -10
Program does not prompt for the validation
set
-10 -10
Program does not prompt for test set -10 -10
Program does not print out weights and J for
the training data.
-15 -15
Program does not print out J for the
validation set
-10 -10
Program does not print out J and Adjusted R2
for the test set
-10 -10
Data not clearly labelled. -10 -10
Late submissions -3
n-1
-3
n-1
I suggest implementing this program using the Normal Equation instead of Gradient Descent, but either are
acceptable.
