5 ClojisR example: Titanic #0
(ns clojisr.v1.tutorials.titanic0
(:require [scicloj.kindly.v4.kind :as kind]
[scicloj.kindly.v4.api :as kindly]))This notebook is a variation of Pradeep Tripathi’s Titanic Kaggle solution in R. Instead of writing it in R as the original, we write it in Clojure, and call R from Clojure.
The goal is to study the Clojure-R interop, and expecially experiment with various ways to define Clojure functions corresponding to R functions.
In this first, naive version, the corresponding Clojure functions are rather simple. They expect a varying number of arguments, and pass those arguments to R function calls by a rather generic way (as defined by Clojisrs)
We do not try to replace the rather imperative style of the original tutorial. Rather, we try to write something that is as close as possible to the original.
We have leanred a lot from this use case. It did expose lots of issues and open questions about the Clojisr API and implementation. Note, however, that the piece of R code that we are mimicing here is not so typical to the current tidyverse trends – there is no heavy use of dplyr, tidy evaluation, etc. It may be a good idea to study other examples that have more of those.
daslu, Jan. 2020
5.1 Bringing the neecessary R functions
Here are most of the functions that we need, brought by the standard require-r mechanism, inspired by libpython-clj’s require-python (though not as sophisticated at the moment). In function names, dots are changed to hyphens.
(require
'[clojisr.v1.r :as r
:refer [r r->clj
r== r!= r< r> r<= r>= r& r&& r| r||
str-md
r+
bra bra<- brabra brabra<- colon
require-r]]
'[clojisr.v1.applications.plotting :refer [plot->svg]]
'[clojure.string :as string]
'[clojure.java.io :as io])(r/set-default-session-type! :rserve){:session-type :rserve}(r/discard-all-sessions){}(require-r
'[base :refer [round names ! set-seed sum which rnorm lapply sapply %in% table list-files c paste colnames row-names cbind gsub <- $ $<- as-data-frame data-frame nlevels factor expression is-na strsplit as-character summary table]]
'[stats :refer [median predict]]
'[ggplot2 :refer [ggsave qplot ggplot aes facet_grid geom_density geom_text geom_histogram geom_bar scale_x_continuous scale_y_continuous labs coord_flip geom_vline geom_hline geom_boxplot]]
'[ggthemes :refer [theme_few]]
'[scales :refer [dollar_format]]
'[graphics :refer [par plot hist dev-off legend]]
'[dplyr :refer [mutate bind_rows summarise group_by]]
'[utils :refer [read-csv write-csv head]]
'[mice :refer [mice complete]]
'[randomForest :refer [randomForest importance]])nil5.2 Introduction – Prediction of Titanic Survival using Random Forest
Pradeep Tripathi’s solution will use randomForest to create a model predicting survival on the Titanic.
5.3 Reading test and train data
This step assumes that the Titanic data lies under the resources/data/ path under your Clojure project.
(def data-path
(-> (io/resource "data/titanic")
(io/file)
(.getAbsolutePath)
(str "/")))# Original code:
list.files('../input')(list-files data-path)[1] "test.csv.gz" "train.csv.gz"
# Original code:
train <-read.csv('../input/train.csv', stringsAsFactors = F)
test <-read.csv('../input/test.csv', stringsAsFactors = F)(def train (read-csv
(str data-path "train.csv.gz")
:stringsAsFactors false))(def test (read-csv
(str data-path "test.csv.gz")
:stringsAsFactors false))5.4 Combining test and train data
As explained by Thripathi, the Random Forest algorithm will use the Bagging method to create multiple random samples with replacement from the dataset, that will be treated as training data, while the out of bag samples will be treated as test data.
# Original code:
titanic<-bind_rows(train,test)(def titanic
(bind_rows train test))5.5 Data check
# Original code:
str(titanic)
summary(titanic)
head(titanic)(kind/md
(str-md titanic))'data.frame': 1309 obs. of 12 variables:
$ PassengerId: int 1 2 3 4 5 6 7 8 9 10 ...
$ Survived : int 0 1 1 1 0 0 0 0 1 1 ...
$ Pclass : int 3 1 3 1 3 3 1 3 3 2 ...
$ Name : chr "Braund, Mr. Owen Harris" "Cumings, Mrs. John Bradley (Florence Briggs Thayer)" "Heikkinen, Miss. Laina" "Futrelle, Mrs. Jacques Heath (Lily May Peel)" ...
$ Sex : chr "male" "female" "female" "female" ...
$ Age : num 22 38 26 35 35 NA 54 2 27 14 ...
$ SibSp : int 1 1 0 1 0 0 0 3 0 1 ...
$ Parch : int 0 0 0 0 0 0 0 1 2 0 ...
$ Ticket : chr "A/5 21171" "PC 17599" "STON/O2. 3101282" "113803" ...
$ Fare : num 7.25 71.28 7.92 53.1 8.05 ...
$ Cabin : chr "" "C85" "" "C123" ...
$ Embarked : chr "S" "C" "S" "S" ...(summary titanic) PassengerId Survived Pclass Name
Min. : 1 Min. :0.0000 Min. :1.000 Length:1309
1st Qu.: 328 1st Qu.:0.0000 1st Qu.:2.000 Class :character
Median : 655 Median :0.0000 Median :3.000 Mode :character
Mean : 655 Mean :0.3838 Mean :2.295
3rd Qu.: 982 3rd Qu.:1.0000 3rd Qu.:3.000
Max. :1309 Max. :1.0000 Max. :3.000
NA's :418
Sex Age SibSp Parch
Length:1309 Min. : 0.17 Min. :0.0000 Min. :0.000
Class :character 1st Qu.:21.00 1st Qu.:0.0000 1st Qu.:0.000
Mode :character Median :28.00 Median :0.0000 Median :0.000
Mean :29.88 Mean :0.4989 Mean :0.385
3rd Qu.:39.00 3rd Qu.:1.0000 3rd Qu.:0.000
Max. :80.00 Max. :8.0000 Max. :9.000
NA's :263
Ticket Fare Cabin Embarked
Length:1309 Min. : 0.000 Length:1309 Length:1309
Class :character 1st Qu.: 7.896 Class :character Class :character
Mode :character Median : 14.454 Mode :character Mode :character
Mean : 33.295
3rd Qu.: 31.275
Max. :512.329
NA's :1
(head titanic) PassengerId Survived Pclass
1 1 0 3
2 2 1 1
3 3 1 3
4 4 1 1
5 5 0 3
6 6 0 3
Name Sex Age SibSp Parch
1 Braund, Mr. Owen Harris male 22 1 0
2 Cumings, Mrs. John Bradley (Florence Briggs Thayer) female 38 1 0
3 Heikkinen, Miss. Laina female 26 0 0
4 Futrelle, Mrs. Jacques Heath (Lily May Peel) female 35 1 0
5 Allen, Mr. William Henry male 35 0 0
6 Moran, Mr. James male NA 0 0
Ticket Fare Cabin Embarked
1 A/5 21171 7.2500 S
2 PC 17599 71.2833 C85 C
3 STON/O2. 3101282 7.9250 S
4 113803 53.1000 C123 S
5 373450 8.0500 S
6 330877 8.4583 Q
Tripathi: We’ve got a sense of our variables, their class type, 3 and the first few observations of each. We know we’re working with 1309 observations of 12 variables.
5.6 Feature engineering
Thripathi’s explanation: We can break down Passenger name into additional meaningful variables which can feed predictions or be used in the creation of additional new variables. For instance, passenger title is contained within the passenger name variable and we can use surname to represent families.
# Original code:
colnames(titanic)(colnames titanic) [1] "PassengerId" "Survived" "Pclass" "Name" "Sex"
[6] "Age" "SibSp" "Parch" "Ticket" "Fare"
[11] "Cabin" "Embarked"
Retrieve title from passenger names
# Original code:
titanic$title<-gsub('(.*, )|(\..*)', '', titanic$Name)(def titanic
($<- titanic 'title
(gsub "(.*, )|(\\\\..*)"
""
($ titanic 'Name))))Show title counts by sex
# Original code:
table(titanic$Sex, titanic$title)Clojisr can covert an R frequency table to a Clojure data structure:
(-> (table ($ titanic 'Sex)
($ titanic 'title))
r->clj)_unnamed [36 3]:
| 0 | 1 | :$value |
|---|---|---|
| female | Capt | 0 |
| female | Col | 1 |
| female | Don | 0 |
| female | Dona | 4 |
| female | Dr | 0 |
| female | Jonkheer | 1 |
| female | Lady | 1 |
| female | Major | 0 |
| female | Master | 1 |
| female | Miss | 7 |
| … | … | … |
| male | Major | 757 |
| male | Master | 197 |
| male | Miss | 0 |
| male | Mlle | 2 |
| male | Mme | 0 |
| male | Mr | 0 |
| male | Mrs | 8 |
| male | Ms | 0 |
| male | Rev | 1 |
| male | Sir | 1 |
| male | the Countess | 0 |
Sometimes, it is convenient to first convert it to an R data frame:
(as-data-frame
(table ($ titanic 'Sex)
($ titanic 'title))) Var1 Var2 Freq
1 female Capt 0
2 male Capt 1
3 female Col 0
4 male Col 4
5 female Don 0
6 male Don 1
7 female Dona 1
8 male Dona 0
9 female Dr 1
10 male Dr 7
11 female Jonkheer 0
12 male Jonkheer 1
13 female Lady 1
14 male Lady 0
15 female Major 0
16 male Major 2
17 female Master 0
18 male Master 61
19 female Miss 260
20 male Miss 0
21 female Mlle 2
22 male Mlle 0
23 female Mme 1
24 male Mme 0
25 female Mr 0
26 male Mr 757
27 female Mrs 197
28 male Mrs 0
29 female Ms 2
30 male Ms 0
31 female Rev 0
32 male Rev 8
33 female Sir 0
34 male Sir 1
35 female the Countess 1
36 male the Countess 0
Sometimes, it is convenient to use the way R prints a frequency table.
(table ($ titanic 'Sex)
($ titanic 'title))
Capt Col Don Dona Dr Jonkheer Lady Major Master Miss Mlle Mme Mr Mrs
female 0 0 0 1 1 0 1 0 0 260 2 1 0 197
male 1 4 1 0 7 1 0 2 61 0 0 0 757 0
Ms Rev Sir the Countess
female 2 0 0 1
male 0 8 1 0
Convert titles with low count into a new title, and rename/reassign Mlle, Ms and Mme.
# Original code:
unusual_title<-c('Dona', 'Lady', 'the Countess','Capt', 'Col', 'Don',
'Dr', 'Major', 'Rev', 'Sir', 'Jonkheer')(def unusual-title
["Dona", "Lady", "the Countess","Capt", "Col", "Don",
"Dr", "Major", "Rev", "Sir", "Jonkheer"])# Original code:
titanic$title[titanic$title=='Mlle']<-'Miss'
titanic$title[titanic$title=='Ms']<-'Miss'
titanic$title[titanic$title=='Mme']<-'Mrs'
titanic$title[titanic$title %in% unusual_title]<-'Unusual Title'(def titanic
(-> titanic
(bra<- (r== ($ titanic 'title) "Mlle")
"title"
"Miss")
(bra<- (r== ($ titanic 'title) "Ms")
"title"
"Miss")
(bra<- (r== ($ titanic 'title) "Mme")
"title"
"Mrs")
(bra<- (%in% ($ titanic 'title) unusual-title)
"title"
"Mrs")))Check the title count again:
# Original code:
table(titanic$Sex, titanic$title)trying again:
(table ($ titanic 'Sex)
($ titanic 'title))
Master Miss Mr Mrs
female 0 264 0 202
male 61 0 757 25
Create a variable which contain the surnames of passengers.
# Original code:
titanic$surname<-sapply(titanic$Name, function(x) strsplit(x,split='[,.]')[[1]][1])
nlevels(factor(titanic$surname)) ## 875 unique surnames(def titanic
($<- titanic 'surname
(sapply ($ titanic 'Name)
(r '(function [x]
(bra (brabra (strsplit x :split "[,.]") 1) 1))))))(-> titanic
($ 'surname)
factor
nlevels)[1] 875
Tripathi: Family size variable: We are going to create a variable “famsize” to know the number of family members. It includes number of sibling/number of parents and children+ passenger themselves
# Original code:
titanic$famsize <- titanic$SibSp + titanic$Parch + 1(def titanic
($<- titanic 'famsize
(r+ ($ titanic 'SibSp)
($ titanic 'Parch)
1)))Create a family variable:
# Original code:
titanic$family <- paste(titanic$surname, titanic$famsize, sep='_')(def titanic
($<- titanic 'family
(paste ($ titanic 'surname)
($ titanic 'famsize)
:sep "_")))Visualize the relationship between family size & survival:
ggplot(titanic[1:891,], aes(x = famsize, fill = factor(Survived))) +
geom_bar(stat='count', position='dodge') +
scale_x_continuous(breaks=c(1:11)) +
labs(x = 'Family Size') +
theme_few()(-> titanic
(bra (colon 1 891)
nil)
(ggplot (aes :x 'famsize
:fill '(factor Survived)))
(r+ (geom_bar :stat "count"
:position "dodge")
(scale_x_continuous :breaks (colon 1 11))
(labs :x "Family Size")
(theme_few))
plot->svg)Tripathi: Explanation: We can see that there’s a survival penalty to single/alone, and those with family sizes above 4. We can collapse this variable into three levels which will be helpful since there are comparatively fewer large families.
Discretize family size:
# Original code:
titanic$fsizeD[titanic$famsize == 1] <- 'single'
titanic$fsizeD[titanic$famsize < 5 & titanic$famsize> 1] <- 'small'
titanic$fsizeD[titanic$famsize> 4] <- 'large'(def titanic
(-> titanic
(bra<- (r== ($ titanic 'famsize) 1)
"fsizeD"
"single")
(bra<- (r& (r< ($ titanic 'famsize) 5)
(r> ($ titanic 'famsize) 1))
"fsizeD"
"small")
(bra<- (r> ($ titanic 'famsize) 4)
"fsizeD" "large")))Let us check if it makes sense:
(-> titanic
($ 'fsizeD)
table)
large single small
82 790 437
And let us make sure there are no missing values:
(-> titanic
($ 'fsizeD)
is-na
table)
FALSE
1309
Tripathi: There’s could be some useful information in the passenger cabin variable including about their deck, so Retrieve deck from Cabin variable.
# Original code:
titanic$Cabin[1:28](-> titanic
(bra (colon 1 28)
"Cabin")) [1] "" "C85" "" "C123" ""
[6] "" "E46" "" "" ""
[11] "G6" "C103" "" "" ""
[16] "" "" "" "" ""
[21] "" "D56" "" "A6" ""
[26] "" "" "C23 C25 C27"
The first character is the deck:
# Original code:
strsplit(titanic$Cabin[2], NULL) [[1]](-> titanic
($ 'Cabin)
(bra 2)
(strsplit nil)
(brabra 1))[1] "C" "8" "5"
Deck variable:
# Original R code:
titanic$deck<-factor(sapply(titanic$Cabin, function(x) strsplit(x, NULL)[[1]][1]))(def titanic
($<- titanic 'deck
(factor (sapply ($ titanic 'Cabin)
'(function [x]
(bra (brabra (strsplit x nil) 1) 1))))))Let us check:
(-> titanic
($ 'deck)
table)
A B C D E F G T
22 65 94 46 41 21 5 1
5.7 Missing values
updated summary
(summary titanic) PassengerId Survived Pclass Name
Min. : 1 Min. :0.0000 Min. :1.000 Length:1309
1st Qu.: 328 1st Qu.:0.0000 1st Qu.:2.000 Class :character
Median : 655 Median :0.0000 Median :3.000 Mode :character
Mean : 655 Mean :0.3838 Mean :2.295
3rd Qu.: 982 3rd Qu.:1.0000 3rd Qu.:3.000
Max. :1309 Max. :1.0000 Max. :3.000
NA's :418
Sex Age SibSp Parch
Length:1309 Min. : 0.17 Min. :0.0000 Min. :0.000
Class :character 1st Qu.:21.00 1st Qu.:0.0000 1st Qu.:0.000
Mode :character Median :28.00 Median :0.0000 Median :0.000
Mean :29.88 Mean :0.4989 Mean :0.385
3rd Qu.:39.00 3rd Qu.:1.0000 3rd Qu.:0.000
Max. :80.00 Max. :8.0000 Max. :9.000
NA's :263
Ticket Fare Cabin Embarked
Length:1309 Min. : 0.000 Length:1309 Length:1309
Class :character 1st Qu.: 7.896 Class :character Class :character
Mode :character Median : 14.454 Mode :character Mode :character
Mean : 33.295
3rd Qu.: 31.275
Max. :512.329
NA's :1
title surname famsize family
Length:1309 Length:1309 Min. : 1.000 Length:1309
Class :character Class :character 1st Qu.: 1.000 Class :character
Mode :character Mode :character Median : 1.000 Mode :character
Mean : 1.884
3rd Qu.: 2.000
Max. :11.000
fsizeD deck
Length:1309 C : 94
Class :character B : 65
Mode :character D : 46
E : 41
A : 22
(Other): 27
NA's :1014
Thripathi’s explanation, following the summary: - Age : 263 missing values - Fare : 1 missing values - Embarked : 2 missing values - survived:too many - Cabin : too many
Missing value in Embarkment – Tripathi: Now we will explore missing values and rectify it through imputation. There are a number of different ways we could go about doing this. Given the small size of the dataset, we probably should not opt for deleting either entire observations (rows) or variables (columns) containing missing values. We’re left with the option of replacing missing values with sensible values given the distribution of the data, e.g., the mean, median or mode.
To know which passengers have no listed embarkment port:
# Original code:
titanic$Embarked[titanic$Embarked == ""] <- NA
titanic[(which(is.na(titanic$Embarked))), 1]Marking as missing:
(def titanic
(bra<- titanic
(r== ($ titanic 'Embarked) "")
"Embarked"
'NA))Checking which has missing port:
(-> titanic
(bra (-> titanic
($ 'Embarked)
is-na
which)
1))[1] 62 830
Tripathi: Passengers 62 and 830 are missing Embarkment.
# Original code:
titanic[c(62, 830), 'Embarked'](-> titanic
(bra [62 830]
"Embarked"))[1] NA NA
Tripathi: So Passenger numbers 62 and 830 are each missing their embarkment ports. Let’s look at their class of ticket and their fare.
# Original code:
titanic[c(62, 830), c(1,3,10)](-> titanic
(bra [62 830]
[1 3 10])) PassengerId Pclass Fare
62 62 1 80
830 830 1 80
Alternatively:
(-> titanic
(bra [62 830]
["PassengerId" "Pclass" "Fare"])) PassengerId Pclass Fare
62 62 1 80
830 830 1 80
Thripathi’s explanation: Both passengers had first class tickets that they spent 80 (pounds?) on. Let’s see the embarkment ports of others who bought similar kinds of tickets.
First way of handling missing value in Embarked:
# Original code:
titanic%>%
group_by(Embarked, Pclass) %>%
filter(Pclass == "1") %>%
filter(Pclass == "1") %>%
filter(Pclass == "1") %>%
summarise(mfare = median(Fare),n = n())(-> titanic
(group_by 'Embarked 'Pclass)
(r.dplyr/filter '(== Pclass "1"))
(summarise :mfare '(median Fare)
:n '(n)))# A tibble: 4 × 4
# Groups: Embarked [4]
Embarked Pclass mfare n
<chr> <int> <dbl> <int>
1 C 1 76.7 141
2 Q 1 90 3
3 S 1 52 177
4 <NA> 1 80 2
Tripathi: Looks like the median price for a first class ticket departing from ‘C’ (Charbourg) was 77 (in comparison to our 80). While first class tickets departing from ‘Q’ were only slightly more expensiive (median price 90), only 3 first class passengers departed from that port. It seems far more likely that passengers 62 and 830 departed with the other 141 first-class passengers from Charbourg.
Second Way of handling missing value in Embarked:
# Original code:
embark_fare <- titanic %>%
filter(PassengerId != 62 & PassengerId != 830)
embark_fare(def embark_fare
(-> titanic
(r.dplyr/filter '(& (!= PassengerId 62)
(!= PassengerId 830)))))Use ggplot2 to visualize embarkment, passenger class, & median fare:
# Original code:
ggplot(embark_fare, aes(x = Embarked, y = Fare, fill = factor(Pclass))) +
geom_boxplot() +
geom_hline(aes(yintercept=80),
colour='red', linetype='dashed', lwd=2) +
scale_y_continuous(labels=dollar_format()) +
theme_few()(-> embark_fare
(ggplot (aes :x 'Embarked
:y 'Fare
:fill '(factor Pclass)))
(r+ (geom_boxplot)
(geom_hline (aes :yintercept 80)
:colour "red"
:linetype "dashed"
:lwd 2)
(scale_y_continuous :labels (dollar_format)))
plot->svg)Tripathi: From plot we can see that The median fare for a first class passenger departing from Charbourg (‘C’) coincides nicely with the $80 paid by our embarkment-deficient passengers. I think we can safely replace the NA values with ‘C’. Since their fare was $80 for 1st class, they most likely embarked from ‘C’.
# Original code:
titanic$Embarked[c(62, 830)] <- 'C'(def titanic
(bra<- titanic [62 830] "Embarked"
"C"))A missing value in fare. Thripathi’s explanation: To know Which passenger has no fare information:
# Original code:
titanic[(which(is.na(titanic$Fare))) , 1](-> titanic
(bra (-> titanic
($ 'Fare)
is-na
which)
1))[1] 1044
Tripathi: Looks like Passenger number 1044 has no listed Fare
Where did this passenger leave from? What was their class?
# Original code:
titanic[1044, c(3, 12)](-> titanic
(bra 1044 [3 12])) Pclass Embarked
1044 3 S
Tripathi: Another way to know about passenger id 1044 :Show row 1044
# Original code:
titanic[1044, ](-> titanic
(bra 1044 nil)) PassengerId Survived Pclass Name Sex Age SibSp Parch
1044 1044 NA 3 Storey, Mr. Thomas male 60.5 0 0
Ticket Fare Cabin Embarked title surname famsize family fsizeD deck
1044 3701 NA S Mr Storey 1 Storey_1 single <NA>
Thripathi’s explanation: Looks like he left from ‘S’ (Southampton) as a 3rd class passenger. Let’s see what other people of the same class and embarkment port paid for their tickets.
6 First way:
titanic%>% filter(Pclass == ‘3’ & Embarked == ‘S’) %>% summarise(missing_fare = median(Fare, na.rm = TRUE))
(-> titanic
(r.dplyr/filter '(& (== Pclass "3")
(== Embarked "S")))
(summarise :missing_fare '(median Fare :na.rm true))) missing_fare
1 8.05
Tripathi: Looks like the median cost for a 3rd class passenger leaving out of Southampton was 8.05. That seems like a logical value for this passenger to have paid.
Second way:
# Original code:
ggplot(titanic[titanic$Pclass == '3' & titanic$Embarked == 'S', ],
aes(x = Fare)) +
geom_density(fill = '#99d6ff', alpha=0.4) +
geom_vline(aes(xintercept=median(Fare, na.rm=T)),
colour='red', linetype='dashed', lwd=1) +
scale_x_continuous(labels=dollar_format()) +
theme_few()(-> titanic
(bra (r& (r== ($ titanic 'Pclass) 3)
(r== ($ titanic 'Embarked) "S"))
nil)
(ggplot (aes :x 'Fare))
(r+ (geom_density :fill "#99d6ff"
:alpha 0.4)
(geom_vline (aes :xintercept
'(median Fare :na.rm true))
:colour "red"
:linetype "dashed"
:lwd 1)
(scale_x_continuous :labels (dollar_format)))
plot->svg)Tripathi: From this visualization, it seems quite reasonable to replace the NA Fare value with median for their class and embarkment which is $8.05.
Replace that NA with 8.05
# Original code:
titanic$Fare[1044] <- 8.05
summary(titanic$Fare)(def titanic
(bra<- titanic 1044 "Fare"
8.05))(-> titanic
($ 'Fare)
summary) Min. 1st Qu. Median Mean 3rd Qu. Max.
0.000 7.896 14.454 33.276 31.275 512.329
Tripathi: Another way of Replace missing fare value with median fare for class/embarkment:
# Original code:
titanic$Fare[1044] <- median(titanic[titanic$Pclass == '3' & titanic$Embarked == 'S', ]$Fare, na.rm = TRUE)(def titanic
(bra<- titanic 1044 "Fare"
(-> titanic
(bra (r& (r== ($ titanic 'Pclass) 3)
(r== ($ titanic 'Embarked) "S"))
"Fare")
(median :na.rm true))))Missing Value in Age.
Tripathi: Show number of missing Age values.
# Original code:
sum(is.na(titanic$Age)) ```
before
::: {.sourceClojure}
```clojure
(-> titanic
($ 'Age)
is-na
sum):::
[1] 263
Tripathi: 263 passengers have no age listed. Taking a median age of all passengers doesn’t seem like the best way to solve this problem, so it may be easiest to try to predict the passengers’ age based on other known information.
To predict missing ages, I’m going to use the mice package. To start with I will factorize the factor variables and then perform mice(multiple imputation using chained equations).
Set a random seed:
# Original code:
set.seed(129)(set-seed 129)NULL
Tripathi: Perform mice imputation, excluding certain less-than-useful variables:
# Original code:
mice_mod <- mice(titanic[, !names(titanic) %in% c('PassengerId','Name','Ticket','Cabin','Family','Surname','Survived')], method='rf')(def mice-mod
(-> titanic
(bra nil
(-> titanic
names
(%in% ["PassengerId","Name","Ticket","Cabin","Family","Surname","Survived"])
!))
(mice :method "rf")))Save the complete output.
# Original code:
mice_output <- complete(mice_mod)(def mice-output
(complete mice-mod))Tripathi: Let’s compare the results we get with the original distribution of passenger ages to ensure that nothing has gone completely awry.
Plot age distributions:
# Original code:
par(mfrow=c(1,2))
hist(titanic$Age, freq=F, main='Age: Original Data',
col='darkred', ylim=c(0,0.04))
hist(mice_output$Age, freq=F, main='Age: MICE Output',
col='lightgreen', ylim=c(0,0.04))(plot->svg
(fn []
(par :mfrow [1 2])
(-> titanic
($ 'Age)
(hist :freq 'F
:main "Age: Original Data"
:col "darkred"
:lim [0 0.04]
:xlab "Age"))
(-> mice-output
($ 'Age)
(hist :freq 'F
:main "Age: MICE Output"
:col "lightgreen"
:lim [0 0.04]
:xlab "Age"))))Tripathi: Things look good, so let’s replace our age vector in the original data with the output from the mice model.
Replace Age variable from the mice model:
# Original code:
titanic$Age <- mice_output$Age
(def titanic
($<- titanic 'Age
($ mice-output 'Age)))Show new number of missing Age values
# Original code:
sum(is.na(titanic$Age))after
(-> titanic
($ 'Age)
is-na
sum)[1] 0
6.1 Feature Enginnering: Part 2
Tripathi: I will create a couple of new age-dependent variables: Child and Mother. A child will simply be someone under 18 years of age and a mother is a passenger who is 1) female, 2) is over 18, 3) has more than 0 children and 4) does not have the title ‘Miss’.
Relationship between age & survival: I include Sex since we know it’s a significant predictor.
# Original code:
ggplot(titanic[1:891,], aes(Age, fill = factor(Survived))) +
geom_histogram() + facet_grid(.~Sex) + theme_few()(-> titanic
(bra (colon 1 891) nil)
(ggplot (aes 'Age :fill '(factor Survived)))
(r+ (geom_histogram)
(facet_grid '(tilde . Sex))
(theme_few))
plot->svg)Tripathi: Create the column Child, and indicate whether child or adult:
# Original code:
titanic$Child[titanic$Age < 18] <- 'Child'
titanic$Child[titanic$Age >= 18] <- 'Adult'(def titanic
(-> titanic
(bra<- (r< ($ titanic 'Age) 18) "Child"
"Child")
(bra<- (r>= ($ titanic 'Age) 18) "Child"
"Adult")))Show counts:
# Original code:
table(titanic$Child, titanic$Survived)(table ($ titanic 'Child)
($ titanic 'Survived))
0 1
Adult 484 273
Child 65 69
Adding Mother variable:
# Original code:
titanic$Mother <- 'Not Mother'
titanic$Mother[titanic$Sex == 'female' & titanic$Parch >0 & titanic$Age > 18 & titanic$title != 'Miss'] <- 'Mother'(def titanic
(-> titanic
($<- 'Mother
"Not Mother")
(bra<- (reduce r&
[(r== ($ titanic 'Sex) "female")
(r> ($ titanic 'Parch) 0)
(r> ($ titanic 'Age) 18)
(r!= ($ titanic 'title) "Miss")])
"Mother"
"Mother")))Show counts:
# Original code:
table(titanic$Mother, titanic$Survived)(table ($ titanic 'Mother)
($ titanic 'Survived))
0 1
Mother 15 38
Not Mother 534 304
Factorizing variables:
# Original code:
titanic$Child <- factor(titanic$Child)
titanic$Mother <- factor(titanic$Mother)
titanic$Pclass <- factor(titanic$Pclass)
titanic$Sex <- factor(titanic$Sex)
titanic$Embarked <- factor(titanic$Embarked)
titanic$Survived <- factor(titanic$Survived)
titanic$title <- factor(titanic$title)
titanic$fsizeD <- factor(titanic$fsizeD)(def titanic
(reduce (fn [data symbol]
($<- data symbol
(factor ($ data symbol))))
titanic
'[Child Mother Pclass Sex Embarked Survived title fsizeD]))Check classes of all columns:
(lapply titanic (r "class"))$PassengerId
[1] "integer"
$Survived
[1] "factor"
$Pclass
[1] "factor"
$Name
[1] "character"
$Sex
[1] "factor"
$Age
[1] "numeric"
$SibSp
[1] "integer"
$Parch
[1] "integer"
$Ticket
[1] "character"
$Fare
[1] "numeric"
$Cabin
[1] "character"
$Embarked
[1] "factor"
$title
[1] "factor"
$surname
[1] "character"
$famsize
[1] "integer"
$family
[1] "character"
$fsizeD
[1] "factor"
$deck
[1] "factor"
$Child
[1] "factor"
$Mother
[1] "factor"
7 Prediction
Split into training & test sets:
# Original code:
train <- titanic[1:891,]
test <- titanic[892:1309,](def train
(bra titanic (colon 1 891) nil))(def test
(bra titanic (colon 892 1309) nil))Building the model:
Tripathi: We then build our model using randomForest on the training set.
Set a random seed:
# Original code:
set.seed(754)(set-seed 754)NULL
Tripathi: Build the model (note: not all possible variables are used):
# Original code:
titanic_model <- randomForest(Survived ~ Pclass + Sex + Age + SibSp + Parch +
Fare + Embarked + title +
fsizeD + Child + Mother,
data = train)(def titanic-model
(randomForest '(tilde Survived
(+ Pclass Sex Age SibSp Parch
Fare Embarked title
fsizeD Child Mother))
:data train))Show model error:
# Original code:
plot(titanic_model, ylim=c(0,0.36))
legend('topright', colnames(titanic_model$err.rate), col=1:3, fill=1:3)(plot->svg
(fn []
(plot titanic-model :ylim [0 0.36]
:main "Model Error")
(legend "topright"
(colnames ($ titanic-model 'err.rate))
:col (colon 1 3)
:fill (colon 1 3))))Tripathi: The black line shows the overall error rate which falls below 20%. The red and green lines show the error rate for ‘died’ and ‘survived’, respectively. We can see that right now we’re much more successful predicting death than we are survival.
7.1 Variable Importance
Get importance:
# Original code:
importance <- importance(titanic_model)
varImportance <- data.frame(Variables = row.names(importance),
Importance = round(importance[ ,'MeanDecreaseGini'],2))(importance titanic-model) MeanDecreaseGini
Pclass 29.479780
Sex 61.243379
Age 46.395186
SibSp 11.921763
Parch 7.785026
Fare 59.766976
Embarked 9.710976
title 65.424749
fsizeD 17.363445
Child 4.306876
Mother 2.240933
(def importance-info
(importance titanic-model))(def var-importance
(data-frame :Variables (row-names importance-info)
:Importance (-> importance-info
(bra nil "MeanDecreaseGini")
round)))importance-info MeanDecreaseGini
Pclass 29.479780
Sex 61.243379
Age 46.395186
SibSp 11.921763
Parch 7.785026
Fare 59.766976
Embarked 9.710976
title 65.424749
fsizeD 17.363445
Child 4.306876
Mother 2.240933
var-importance Variables Importance
Pclass Pclass 29
Sex Sex 61
Age Age 46
SibSp SibSp 12
Parch Parch 8
Fare Fare 60
Embarked Embarked 10
title title 65
fsizeD fsizeD 17
Child Child 4
Mother Mother 2
7.2 Variable importance
Create a rank variable based on importance:
# Original code:
rankImportance <- varImportance %>%
mutate(Rank = paste0('#',dense_rank(desc(Importance))))(def rank-importance
(-> var-importance
(mutate :Rank '(paste0 "#" (dense_rank (desc Importance))))))rank-importance Variables Importance Rank
Pclass Pclass 29 #5
Sex Sex 61 #2
Age Age 46 #4
SibSp SibSp 12 #7
Parch Parch 8 #9
Fare Fare 60 #3
Embarked Embarked 10 #8
title title 65 #1
fsizeD fsizeD 17 #6
Child Child 4 #10
Mother Mother 2 #11
Tripathi: Use ggplot2 to visualize the relative importance of variables
# Original code:
ggplot(rankImportance, aes(x = reorder(Variables, Importance),
y = Importance, fill = Importance)) +
geom_bar(stat='identity') +
geom_text(aes(x = Variables, y = 0.5, label = Rank),
hjust=0, vjust=0.55, size = 4, colour = 'red') +
labs(x = 'Variables') +
coord_flip() +
theme_few()(-> rank-importance
(ggplot (aes :x '(reorder Variables Importance)
:y 'Importance
:fill 'Importance))
(r+ (geom_bar :stat "Identity")
(geom_text (aes :x 'Variables
:y 0.5
:label 'Rank)
:hjust 0
:vjust 0.55
:size 4
:colour "red")
(labs :x "Variables")
(coord_flip)
(theme_few))
plot->svg)Tripathi: From the plot we can see that the ‘title’ variable has the highest relative importance out of all of our predictor variables.
7.3 Final Prediction
Predict using the test set:
# Original code:
prediction <- predict(titanic_model, test)
prediction(def prediction
(predict titanic-model test))Tripathi: Save the solution to a dataframe with two columns: PassengerId and Survived (prediction).
# Original code:
Output<- data.frame(PassengerID = test$PassengerId, Survived = prediction)
Output(def output (data-frame :PassengerId ($ test 'PassengerId)
:Survived prediction))(r->clj output)_unnamed [418 3]:
| :$row.names | :PassengerId | :Survived |
|---|---|---|
| 892 | 892 | :0 |
| 893 | 893 | :0 |
| 894 | 894 | :0 |
| 895 | 895 | :0 |
| 896 | 896 | :0 |
| 897 | 897 | :0 |
| 898 | 898 | :1 |
| 899 | 899 | :0 |
| 900 | 900 | :1 |
| 901 | 901 | :0 |
| … | … | … |
| 1299 | 1299 | :0 |
| 1300 | 1300 | :1 |
| 1301 | 1301 | :1 |
| 1302 | 1302 | :1 |
| 1303 | 1303 | :1 |
| 1304 | 1304 | :0 |
| 1305 | 1305 | :0 |
| 1306 | 1306 | :1 |
| 1307 | 1307 | :0 |
| 1308 | 1308 | :0 |
| 1309 | 1309 | :1 |
Write the Output to file:
# Original code:
write.csv(Output, file = 'pradeep_titanic_output.csv', row.names = F)(write-csv output
:file "/tmp/pradeep_titanic_output.csv"
:row.names 'F)NULL
7.4 Conclusion
Tripathi: Thank you for taking the time to read through my first exploration of a Titanic Kaggle dataset. Again, this newbie welcomes comments and suggestions!