(require '[huri.core :as h]
'[clojure.core.matrix.dataset :as d]
'[incanter.core :as i])
(def ds [{:id 1.0 :name "name1"}
{:id 2.0 :name "name2"}
{:id 3.0 :name "name3"}])
;; UPDATE THE :NAME COLUMN IN THE DATASET
;; - IF THE VALUE IS NOT "NAME2", THEN CHANGE TO "NOT 2"
;;
;; EXPECTED OUTPUT:
;; | :id | :name |
;; |-----+-------|
;; | 1.0 | not 2 |
;; | 2.0 | name2 |
;; | 3.0 | not 2 |
;; WITH CLOJURE.CORE/UPDATE
(def d1 (map (fn [x] (update x :name #(if (= "name2" %) % "not 2"))) ds))
;; WITH CLOJURE.CORE/UPDATE-IN
(def d2 (map (fn [x] (update-in x [:name] #(if (= "name2" %) % "not 2"))) ds))
;; WITH HURI/UPDATE-COLS
(def d3 (h/update-cols {:name #(if (= "name2" %) % "not 2")} ds))
;; WITH MATRIX.DATASET/EMAP-COLUMN
(def d4 (-> ds
(d/dataset)
(d/emap-column :name #(if (= "name2" %) % "not 2"))
((comp #(map into %) d/row-maps))))
;; WITH INCANTER/TRANSFORM-COL
(def d5 (-> ds
(i/to-dataset)
(i/transform-col :name #(if (= "name2" %) % "not 2"))
((comp #(map into %) second vals))))
incanter
Adding New Columns to Clojure Map
(require '[huri.core :as h]
'[clojure.core.matrix.dataset :as d]
'[incanter.core :as i])
(def ds [{:id 1.0 :name "name1"}
{:id 2.0 :name "name2"}
{:id 3.0 :name "name3"}])
;; ADD 2 COLUMNS TO THE DATASET
;; - ADD 2 TO ID AND NAME ADD2
;; - CHECK NAME = "name2" AND NAME NAME2
;;
;; EXPECTED OUTPUT:
;;| :id | :name | :add2 | :name2 |
;;|-----+-------+-------+--------|
;;| 1.0 | name1 | 3.0 | N |
;;| 2.0 | name2 | 4.0 | Y |
;;| 3.0 | name3 | 5.0 | N |
;; WITH PLAIN CLOJURE
;; #1 - MERGE
(def d1 (map #(merge % {:add2 (+ (:id %) 2)
:name2 (if (= "name2" (:name %)) "Y" "N")}) ds))
;; #2 - MERGE-WITH
(def d2 (map #(merge-with into % {:add2 (+ (:id %) 2)
:name2 (if (= "name2" (:name %)) "Y" "N")}) ds))
;; #3 - ASSOC
(def d3 (map #(assoc % :add2 (+ (:id %) 2)
:name2 (if (= "name2" (:name %)) "Y" "N")) ds))
;; #4 - CONJ
(def d4 (map #(conj % {:add2 (+ (:id %) 2)
:name2 (if (= "name2" (:name %)) "Y" "N")}) ds))
;; #5 - CONCAT
(def d5 (map #(into {} (concat % {:add2 (+ (:id %) 2)
:name2 (if (= "name2" (:name %)) "Y" "N")})) ds))
;; WITH HURI
(def d6 (h/derive-cols {:name2 [#(if (= "name2" %) "Y" "N") :name]
:add2 [#(+ 2 %) :id]} ds))
;; WITH CORE.MATRIX API
(def d7 (-> ds
(d/dataset)
(d/add-column :add2 (map #(+ 2 %) (map :id ds)))
(d/add-column :name2 (map #(if (= "name2" %) "Y" "N") (map :name ds)))
(d/row-maps)))
;; WITH INCANTER API
(def d8 (->> ds
(i/to-dataset)
(i/add-derived-column :add2 [:id] #(+ 2 %))
(i/add-derived-column :name2 [:name] #(if (= "name2" %) "Y" "N"))
((comp second vals))))
;; CHECK THE DATA EQUALITY
(= d1 d2 d3 d4 d5 d6 d7 d8)
;; true
Transpose in Clojure
(require '[huri.core :as h]
'[clojure.core.matrix.dataset :as d]
'[incanter.core :as i])
;; FROM MAP OF ROWS TO MAP OF COLUMNS
(def byRow [{:x 1 :y "a"}
{:x 2 :y "b"}
{:x 3 :y "c"}])
;; APPROACH #1 - PLAIN CLOJURE
(zipmap (keys (first byRow)) (apply map list (map vals byRow)))
; {:x (1 2 3), :y ("a" "b" "c")}
;; APPROACH #2 - HURI LIBRARY
(h/col-oriented byRow)
; {:x (1 2 3), :y ("a" "b" "c")}
;; APPROACH #3 - CORE.MATRIX LIBRARY
(d/to-map (d/dataset (keys (first byRow)) byRow))
; {:x [1 2 3], :y ["a" "b" "c"]}
;; APPROACH #4 - INCANTER LIBRARY
(i/to-map (i/to-dataset byRow))
; {:x (1 2 3), :y ("a" "b" "c")}
;; FROM MAP OF COLUMNS TO MAP OF ROWS
(def byCol {:x '(1 2 3)
:y '("a" "b" "c")})
;; APPROACH #1 - PLAIN CLOJURE
(map #(zipmap (keys byCol) %) (apply map list (vals byCol)))
; ({:x 1, :y "a"} {:x 2, :y "b"} {:x 3, :y "c"})
;; APPROACH #2 - HURI LIBRARY
(h/row-oriented byCol)
; ({:x 1, :y "a"} {:x 2, :y "b"} {:x 3, :y "c"})
;; APPROACH #3 - CORE.MATRIX LIBRARY
(d/row-maps (d/dataset (keys byCol) byCol))
; [{:x 1, :y "a"} {:x 2, :y "b"} {:x 3, :y "c"}]
;; APPROACH #4 - INCANTER LIBRARY
(second (vals (i/dataset (keys byCol) (apply map list (vals byCol)))))
; ({:x 1, :y "a"} {:x 2, :y "b"} {:x 3, :y "c"})
Two Ways to Select Rows in Incanter
user=> (use '(incanter core io))
nil
user=> (def iris (read-dataset "../data/iris.dat" :header true :delim \space))
#'user/iris
user=> (sel iris :rows (range 3))
[:Sepal.Length :Sepal.Width :Petal.Length :Petal.Width :Species]
[5.1 3.5 1.4 0.2 "setosa"]
[4.9 3.0 1.4 0.2 "setosa"]
[4.7 3.2 1.3 0.2 "setosa"]
;; METHOD 1 - USING $WHERE
user=> ($where {:Species {:in #{"virginica" "setosa"}} :Sepal.Length {:gt 5.5, :lt 6.0}} iris)
[:Sepal.Length :Sepal.Width :Petal.Length :Petal.Width :Species]
[5.8 4.0 1.2 0.2 "setosa"]
[5.7 4.4 1.5 0.4 "setosa"]
[5.7 3.8 1.7 0.3 "setosa"]
[5.8 2.7 5.1 1.9 "virginica"]
[5.7 2.5 5.0 2.0 "virginica"]
[5.8 2.8 5.1 2.4 "virginica"]
[5.6 2.8 4.9 2.0 "virginica"]
[5.8 2.7 5.1 1.9 "virginica"]
[5.9 3.0 5.1 1.8 "virginica"]
;; METHOD 2 - USING QUERY-DATASET
user=> (query-dataset iris {:Species {:in #{"virginica" "setosa"}} :Sepal.Length {:gt 5.5, :lt 6.0}})
[:Sepal.Length :Sepal.Width :Petal.Length :Petal.Width :Species]
[5.8 4.0 1.2 0.2 "setosa"]
[5.7 4.4 1.5 0.4 "setosa"]
[5.7 3.8 1.7 0.3 "setosa"]
[5.8 2.7 5.1 1.9 "virginica"]
[5.7 2.5 5.0 2.0 "virginica"]
[5.8 2.8 5.1 2.4 "virginica"]
[5.6 2.8 4.9 2.0 "virginica"]
[5.8 2.7 5.1 1.9 "virginica"]
[5.9 3.0 5.1 1.8 "virginica"]
Read CSV Data File with Clojure
;; LOAD PACKAGES user=> (use '(incanter core io)) nil ;; READ CSV FILE user=> (def ds1 (read-dataset "../data/credit_count.csv" :header true :delim \,)) #'user/ds1 ;; EXAMINE THE DATA user=> ($ (range 0 3) :all ds1) [:CARDHLDR :DEFAULT :AGE :ACADMOS :ADEPCNT :MAJORDRG :MINORDRG :OWNRENT :INCOME :SELFEMPL :INCPER :EXP_INC :SPENDING :LOGSPEND ] [0 0 27.25 4 0 0 0 0 1200 0 18000 6.667E-4 " " " "] [0 0 40.8333321 111 3 0 0 1 4000 0 13500 2.222E-4 " " " "] [1 0 37.6666679 54 3 0 0 1 3666.6666667 0 11300 0.0332699 121.9896773 4.8039364] ;; CALCULATE SUMMARY BY GROUP user=> ($rollup :mean :INCOME [:CARDHLDR :DEFAULT] ds1) [:DEFAULT :CARDHLDR :INCOME] [1 1 2156.117553547691] [0 1 2653.2908642884945] [0 0 2165.1530843234673]
Read Space-Delimited File with Clojure
;; load clojure libraries user=> (use 'incanter.core 'incanter.io) nil ;; import data file user=> (def iris (read-dataset "../data/iris.dat" :header true :delim \space)) #'user/iris ;; show data header user=> (col-names iris) [:Sepal.Length :Sepal.Width :Petal.Length :Petal.Width :Species] ;; show first 3 data records with $ user=> ($ (range 0 3) :all iris) [:Sepal.Length :Sepal.Width :Petal.Length :Petal.Width :Species] [5.1 3.5 1.4 0.2 "setosa"] [4.9 3.0 1.4 0.2 "setosa"] [4.7 3.2 1.3 0.2 "setosa"]
Yet Another Blog in Statistical Computing 