From 30c616af04d56391a5810b93df825d58017c39de Mon Sep 17 00:00:00 2001
From: Joshua-Dias-Barreto <joshuadiasbarreto@gmail.com>
Date: Mon, 15 May 2023 11:34:32 +0530
Subject: [PATCH 1/2] Added describe method for numeric dataframes

---
 src/DataFrame-Tests/DataFrameTest.class.st | 20 ++++++++++
 src/DataFrame/DataFrame.class.st           | 45 ++++++++++++++++++++++
 2 files changed, 65 insertions(+)

diff --git a/src/DataFrame-Tests/DataFrameTest.class.st b/src/DataFrame-Tests/DataFrameTest.class.st
index 5751ce49..dafa8ee6 100644
--- a/src/DataFrame-Tests/DataFrameTest.class.st
+++ b/src/DataFrame-Tests/DataFrameTest.class.st
@@ -1553,6 +1553,26 @@ DataFrameTest >> testDataTypesWithNil [
 	self assert: df dataTypes equals: expected
 ]
 
+{ #category : #tests }
+DataFrameTest >> testDescribe [
+
+	| dataFrame expected |
+	dataFrame := DataFrame
+		             withRows: #( #( 1 1 ) #( 2 nil ) #( 3 1 ) )
+		             columnNames: #( 'A' 'B' ).
+
+	expected := DataFrame withRows:
+		            #( #( 3 2 1 1 1 2 3 3 SmallInteger )
+		               #( 2 1 0 1 1 1 1 1 SmallInteger ) ).
+
+	expected rowNames: #( 'A' 'B' ).
+	expected columnNames:
+		#( 'count' 'mean' 'std' 'min' '25%' '50%' '75%' 'max' 'dtype' ).
+	expected at: 1 at: 9 put: SmallInteger.
+	expected at: 2 at: 9 put: SmallInteger.
+	self assert: dataFrame describe equals: expected
+]
+
 { #category : #tests }
 DataFrameTest >> testDetect [
 	| actual expected |
diff --git a/src/DataFrame/DataFrame.class.st b/src/DataFrame/DataFrame.class.st
index 73dfb830..d7f6e126 100644
--- a/src/DataFrame/DataFrame.class.st
+++ b/src/DataFrame/DataFrame.class.st
@@ -942,6 +942,51 @@ DataFrame >> defaultHeadTailSize [
 	^ 5
 ]
 
+{ #category : #statistics }
+DataFrame >> describe [
+	"method to statistically describe a numerical dataframe"
+
+	| nCol nRow describeDF col count dtype |
+	nCol := self numberOfColumns.
+	nRow := self numberOfRows.
+	describeDF := self class new: nCol @ 9.
+	describeDF columnNames:
+		#( 'count' 'mean' 'std' 'min' '25%' '50%' '75%' 'max' 'dtype' ).
+	describeDF rowNames: self columnNames.
+	1 to: nCol do: [ :i |
+		| mean std mini fQ sQ tQ maxi |
+		col := self columnAt: i.
+		count := col countNonNils.
+		count = 0 ifFalse: [
+			col := col removeNils.
+			mean := col average.
+			std := col stdev.
+			mini := col min.
+			fQ := col firstQuartile.
+			sQ := col secondQuartile.
+			tQ := col thirdQuartile.
+			maxi := col max ].
+		dtype := col calculateDataType.
+		describeDF at: i at: 1 put: count.
+
+		describeDF at: i at: 2 put: mean.
+
+		describeDF at: i at: 3 put: std.
+
+		describeDF at: i at: 4 put: mini.
+
+		describeDF at: i at: 5 put: fQ.
+
+		describeDF at: i at: 6 put: sQ.
+
+		describeDF at: i at: 7 put: tQ.
+
+		describeDF at: i at: 8 put: maxi.
+
+		describeDF at: i at: 9 put: dtype ].
+	^ describeDF
+]
+
 { #category : #accessing }
 DataFrame >> dimensions [
 	"Returns the number of rows and number of columns in a DataFrame"

From e13f4d5b6f25e4f59905359aa1787c0929635d9c Mon Sep 17 00:00:00 2001
From: Joshua-Dias-Barreto <joshuadiasbarreto@gmail.com>
Date: Thu, 25 May 2023 02:15:36 +0530
Subject: [PATCH 2/2] Implemented a cleaner and more efficient describe method.

---
 src/DataFrame/DataFrame.class.st | 60 ++++++++++----------------------
 1 file changed, 19 insertions(+), 41 deletions(-)

diff --git a/src/DataFrame/DataFrame.class.st b/src/DataFrame/DataFrame.class.st
index d26aab62..dbc5aa20 100644
--- a/src/DataFrame/DataFrame.class.st
+++ b/src/DataFrame/DataFrame.class.st
@@ -943,47 +943,25 @@ DataFrame >> defaultHeadTailSize [
 
 { #category : #statistics }
 DataFrame >> describe [
-	"method to statistically describe a numerical dataframe"
-
-	| nCol nRow describeDF col count dtype |
-	nCol := self numberOfColumns.
-	nRow := self numberOfRows.
-	describeDF := self class new: nCol @ 9.
-	describeDF columnNames:
-		#( 'count' 'mean' 'std' 'min' '25%' '50%' '75%' 'max' 'dtype' ).
-	describeDF rowNames: self columnNames.
-	1 to: nCol do: [ :i |
-		| mean std mini fQ sQ tQ maxi |
-		col := self columnAt: i.
-		count := col countNonNils.
-		count = 0 ifFalse: [
-			col := col removeNils.
-			mean := col average.
-			std := col stdev.
-			mini := col min.
-			fQ := col firstQuartile.
-			sQ := col secondQuartile.
-			tQ := col thirdQuartile.
-			maxi := col max ].
-		dtype := col calculateDataType.
-		describeDF at: i at: 1 put: count.
-
-		describeDF at: i at: 2 put: mean.
-
-		describeDF at: i at: 3 put: std.
-
-		describeDF at: i at: 4 put: mini.
-
-		describeDF at: i at: 5 put: fQ.
-
-		describeDF at: i at: 6 put: sQ.
-
-		describeDF at: i at: 7 put: tQ.
-
-		describeDF at: i at: 8 put: maxi.
-
-		describeDF at: i at: 9 put: dtype ].
-	^ describeDF
+	"Answer another data frame with statistics describing the columns of this data frame"
+
+	| content |
+	content := self numericalColumns collect: [ :column |
+		           {
+			           column countNonNils.
+			           column average.
+			           column stdev.
+			           column min.
+			           column firstQuartile.
+			           column secondQuartile.
+			           column thirdQuartile.
+			           column max.
+			           column calculateDataType } ].
+
+	^ self class
+		  withRows: content
+		  rowNames: self numericalColumnNames
+		  columnNames: #( count mean std min '25%' '50%' '75%' max dtype )
 ]
 
 { #category : #accessing }