steve
/
factor
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

machine-learning.data-sets: renaming slots in the data-set tuple 

"features" is a better name for the slot than "data"
windows-high-dpi
Björn Lindqvist 2018-02-25 15:24:53 +01:00
parent b672ed91fb
commit fa2a0dc779
2 changed files with 16 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
extra/machine-learning/data-sets/data-sets.factor
View File

@ -6,8 +6,10 @@ kernel math.parser sequences splitting ;
IN: machine-learning.data-sets IN: machine-learning.data-sets
TUPLE: data-set data target target-names description TUPLE: data-set
feature-names ; features targets
feature-names target-names
description ;
C: <data-set> data-set C: <data-set> data-set
@ -37,23 +39,25 @@ PRIVATE>
! Omits the identifiers which are not so interesting. ! Omits the identifiers which are not so interesting.
[ but-last [ string>number ] map ] map [ but-last [ string>number ] map ] map
[ [ rest ] map ] [ [ first ] map ] bi [ [ rest ] map ] [ [ first ] map ] bi
{ "a1" "a2" "a3" "a4" "a5" "a6" }
{ "no" "yes" } { "no" "yes" }
"monks.names" load-file "monks.names" load-file
{ "a1" "a2" "a3" "a4" "a5" "a6" } <data-set> ; <data-set> ;
: load-iris ( -- data-set ) : load-iris ( -- data-set )
"iris.csv" load-table-csv "iris.csv" load-table-csv
[ [ unclip-last ] { } map>assoc unzip ] [ 2 tail ] bi* [ [ unclip-last ] { } map>assoc unzip ] [ 2 tail ] bi*
"iris.rst" load-file
{ {
"sepal length (cm)" "sepal width (cm)" "sepal length (cm)" "sepal width (cm)"
"petal length (cm)" "petal width (cm)" "petal length (cm)" "petal width (cm)"
} <data-set> ; } swap
"iris.rst" load-file
<data-set> ;
: load-linnerud ( -- data-set ) : load-linnerud ( -- data-set )
data-set new data-set new
"linnerud_exercise.csv" load-table "linnerud_exercise.csv" load-table
[ >>data ] [ >>feature-names ] bi* [ >>features ] [ >>feature-names ] bi*
"linnerud_physiological.csv" load-table "linnerud_physiological.csv" load-table
[ >>target ] [ >>target-names ] bi* [ >>targets ] [ >>target-names ] bi*
"linnerud.rst" load-file >>description ; "linnerud.rst" load-file >>description ;

10
extra/machine-learning/decision-trees/decision-trees.factor
View File

@ -20,12 +20,12 @@ IN: machine-learning.decision-trees
[ [ entropy2 ] [ length ] bi * ] map-sum ; inline [ [ entropy2 ] [ length ] bi * ] map-sum ; inline
:: average-gain ( dataset idx -- gain ) :: average-gain ( dataset idx -- gain )
dataset target>> :> target dataset targets>> :> targets
dataset data>> :> data dataset features>> :> features
data target zip :> data-target features targets zip :> features-targets
data-target idx subsets-weighted-entropy :> weighted features-targets idx subsets-weighted-entropy :> weighted
target entropy2 weighted data length / - ; targets entropy2 weighted features length / - ;
: highest-gain-index ( dataset -- idx ) : highest-gain-index ( dataset -- idx )
dup feature-names>> length <iota> [ dup feature-names>> length <iota> [