Merge pull request #76 from amitkgupta/master

Improve ConfusionMatrix GetSummary formatting, and several tiny improvements

base/bag_test.go

@@ -35,7 +35,7 @@ func TestBAGSimple(t *testing.T) {
 			} else if name == "2" {
 				attrSpecs[2] = a
 			} else {
-				panic(name)
+				t.Fatalf("Unexpected attribute name '%s'", name)
 			}
 		}
 
@@ -102,7 +102,7 @@ func TestBAG(t *testing.T) {
 			} else if name == "2" {
 				attrSpecs[2] = a
 			} else {
-				panic(name)
+				t.Fatalf("Unexpected attribute name '%s'", name)
 			}
 		}
 

base/csv.go

@@ -11,10 +11,10 @@ import (
 )
 
 // ParseCSVGetRows returns the number of rows in a given file.
-func ParseCSVGetRows(filepath string) int {
+func ParseCSVGetRows(filepath string) (int, error) {
 	file, err := os.Open(filepath)
 	if err != nil {
-		panic(err)
+		return 0, err
 	}
 	defer file.Close()
 
@@ -25,11 +25,11 @@ func ParseCSVGetRows(filepath string) int {
 		if err == io.EOF {
 			break
 		} else if err != nil {
-			panic(err)
+			return 0, err
 		}
 		counter++
 	}
-	return counter
+	return counter, nil
 }
 
 // ParseCSVGetAttributes returns an ordered slice of appropriate-ly typed
@@ -157,7 +157,11 @@ func ParseCSVBuildInstances(filepath string, hasHeaders bool, u UpdatableDataGri
 func ParseCSVToInstances(filepath string, hasHeaders bool) (instances *DenseInstances, err error) {
 
 	// Read the number of rows in the file
-	rowCount := ParseCSVGetRows(filepath)
+	rowCount, err := ParseCSVGetRows(filepath)
+	if err != nil {
+		return nil, err
+	}
+
 	if hasHeaders {
 		rowCount--
 	}
@@ -176,7 +180,7 @@ func ParseCSVToInstances(filepath string, hasHeaders bool) (instances *DenseInst
 	// Read the input
 	file, err := os.Open(filepath)
 	if err != nil {
-		panic(err)
+		return nil, err
 	}
 	defer file.Close()
 	reader := csv.NewReader(file)
@@ -188,7 +192,7 @@ func ParseCSVToInstances(filepath string, hasHeaders bool) (instances *DenseInst
 		if err == io.EOF {
 			break
 		} else if err != nil {
-			panic(err)
+			return nil, err
 		}
 		if rowCounter == 0 {
 			if hasHeaders {

base/csv_test.go

@@ -4,78 +4,92 @@ import (
 	"testing"
 )
 
-func TestParseCSVGetRows(testEnv *testing.T) {
-	lineCount := ParseCSVGetRows("../examples/datasets/iris.csv")
+func TestParseCSVGetRows(t *testing.T) {
+	lineCount, err := ParseCSVGetRows("../examples/datasets/iris.csv")
+	if err != nil {
+		t.Fatalf("Unable to parse CSV to get number of rows: %s", err.Error())
+	}
 	if lineCount != 150 {
-		testEnv.Errorf("Should have %d lines, has %d", 150, lineCount)
+		t.Errorf("Should have %d lines, has %d", 150, lineCount)
+	}
+
+	lineCount, err = ParseCSVGetRows("../examples/datasets/iris_headers.csv")
+	if err != nil {
+		t.Fatalf("Unable to parse CSV to get number of rows: %s", err.Error())
 	}
 
-	lineCount = ParseCSVGetRows("../examples/datasets/iris_headers.csv")
 	if lineCount != 151 {
-		testEnv.Errorf("Should have %d lines, has %d", 151, lineCount)
+		t.Errorf("Should have %d lines, has %d", 151, lineCount)
 	}
 
 }
 
-func TestParseCCSVGetAttributes(testEnv *testing.T) {
+func TestParseCSVGetRowsWithMissingFile(t *testing.T) {
+	_, err := ParseCSVGetRows("../examples/datasets/non-existent.csv")
+	if err == nil {
+		t.Fatal("Expected ParseCSVGetRows to return error when given path to non-existent file")
+	}
+}
+
+func TestParseCCSVGetAttributes(t *testing.T) {
 	attrs := ParseCSVGetAttributes("../examples/datasets/iris_headers.csv", true)
 	if attrs[0].GetType() != Float64Type {
-		testEnv.Errorf("First attribute should be a float, %s", attrs[0])
+		t.Errorf("First attribute should be a float, %s", attrs[0])
 	}
 	if attrs[0].GetName() != "Sepal length" {
-		testEnv.Errorf(attrs[0].GetName())
+		t.Errorf(attrs[0].GetName())
 	}
 
 	if attrs[4].GetType() != CategoricalType {
-		testEnv.Errorf("Final attribute should be categorical, %s", attrs[4])
+		t.Errorf("Final attribute should be categorical, %s", attrs[4])
 	}
 	if attrs[4].GetName() != "Species" {
-		testEnv.Error(attrs[4])
+		t.Error(attrs[4])
 	}
 }
 
-func TestParseCsvSniffAttributeTypes(testEnv *testing.T) {
+func TestParseCsvSniffAttributeTypes(t *testing.T) {
 	attrs := ParseCSVSniffAttributeTypes("../examples/datasets/iris_headers.csv", true)
 	if attrs[0].GetType() != Float64Type {
-		testEnv.Errorf("First attribute should be a float, %s", attrs[0])
+		t.Errorf("First attribute should be a float, %s", attrs[0])
 	}
 	if attrs[1].GetType() != Float64Type {
-		testEnv.Errorf("Second attribute should be a float, %s", attrs[1])
+		t.Errorf("Second attribute should be a float, %s", attrs[1])
 	}
 	if attrs[2].GetType() != Float64Type {
-		testEnv.Errorf("Third attribute should be a float, %s", attrs[2])
+		t.Errorf("Third attribute should be a float, %s", attrs[2])
 	}
 	if attrs[3].GetType() != Float64Type {
-		testEnv.Errorf("Fourth attribute should be a float, %s", attrs[3])
+		t.Errorf("Fourth attribute should be a float, %s", attrs[3])
 	}
 	if attrs[4].GetType() != CategoricalType {
-		testEnv.Errorf("Final attribute should be categorical, %s", attrs[4])
+		t.Errorf("Final attribute should be categorical, %s", attrs[4])
 	}
 }
 
-func TestParseCSVSniffAttributeNamesWithHeaders(testEnv *testing.T) {
+func TestParseCSVSniffAttributeNamesWithHeaders(t *testing.T) {
 	attrs := ParseCSVSniffAttributeNames("../examples/datasets/iris_headers.csv", true)
 	if attrs[0] != "Sepal length" {
-		testEnv.Error(attrs[0])
+		t.Error(attrs[0])
 	}
 	if attrs[1] != "Sepal width" {
-		testEnv.Error(attrs[1])
+		t.Error(attrs[1])
 	}
 	if attrs[2] != "Petal length" {
-		testEnv.Error(attrs[2])
+		t.Error(attrs[2])
 	}
 	if attrs[3] != "Petal width" {
-		testEnv.Error(attrs[3])
+		t.Error(attrs[3])
 	}
 	if attrs[4] != "Species" {
-		testEnv.Error(attrs[4])
+		t.Error(attrs[4])
 	}
 }
 
-func TestReadInstances(testEnv *testing.T) {
+func TestParseCSVToInstances(t *testing.T) {
 	inst, err := ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 	row1 := inst.RowString(0)
@@ -83,27 +97,34 @@ func TestReadInstances(testEnv *testing.T) {
 	row3 := inst.RowString(100)
 
 	if row1 != "5.10 3.50 1.40 0.20 Iris-setosa" {
-		testEnv.Error(row1)
+		t.Error(row1)
 	}
 	if row2 != "7.00 3.20 4.70 1.40 Iris-versicolor" {
-		testEnv.Error(row2)
+		t.Error(row2)
 	}
 	if row3 != "6.30 3.30 6.00 2.50 Iris-virginica" {
-		testEnv.Error(row3)
+		t.Error(row3)
 	}
 }
 
-func TestReadAwkwardInsatnces(testEnv *testing.T) {
+func TestParseCSVToInstancesWithMissingFile(t *testing.T) {
+	_, err := ParseCSVToInstances("../examples/datasets/non-existent.csv", true)
+	if err == nil {
+		t.Fatal("Expected ParseCSVToInstances to return error when given path to non-existent file")
+	}
+}
+
+func TestReadAwkwardInsatnces(t *testing.T) {
 	inst, err := ParseCSVToInstances("../examples/datasets/chim.csv", true)
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 	attrs := inst.AllAttributes()
 	if attrs[0].GetType() != Float64Type {
-		testEnv.Error("Should be float!")
+		t.Error("Should be float!")
 	}
 	if attrs[1].GetType() != CategoricalType {
-		testEnv.Error("Should be discrete!")
+		t.Error("Should be discrete!")
 	}
 }

base/edf/alloc_test.go

@@ -11,18 +11,18 @@ func TestAllocFixed(t *testing.T) {
 	Convey("Creating a non-existent file should succeed", t, func() {
 		tempFile, err := ioutil.TempFile(os.TempDir(), "TestFileCreate")
 		So(err, ShouldEqual, nil)
-		Convey("Mapping the file should suceed", func() {
+		Convey("Mapping the file should succeed", func() {
 			mapping, err := EdfMap(tempFile, EDF_CREATE)
 			So(err, ShouldEqual, nil)
-			Convey("Allocation should suceed", func() {
+			Convey("Allocation should succeed", func() {
 				r, err := mapping.AllocPages(1, 2)
 				So(err, ShouldEqual, nil)
 				So(r.Start.Byte, ShouldEqual, 4*os.Getpagesize())
 				So(r.Start.Segment, ShouldEqual, 0)
-				Convey("Unmapping the file should suceed", func() {
+				Convey("Unmapping the file should succeed", func() {
 					err = mapping.Unmap(EDF_UNMAP_SYNC)
 					So(err, ShouldEqual, nil)
-					Convey("Remapping the file should suceed", func() {
+					Convey("Remapping the file should succeed", func() {
 						mapping, err = EdfMap(tempFile, EDF_READ_ONLY)
 						Convey("Should get the same allocations back", func() {
 							rr, err := mapping.GetThreadBlocks(2)
@@ -41,20 +41,20 @@ func TestAllocWithExtraContentsBlock(t *testing.T) {
 	Convey("Creating a non-existent file should succeed", t, func() {
 		tempFile, err := ioutil.TempFile(os.TempDir(), "TestFileCreate")
 		So(err, ShouldEqual, nil)
-		Convey("Mapping the file should suceed", func() {
+		Convey("Mapping the file should succeed", func() {
 			mapping, err := EdfMap(tempFile, EDF_CREATE)
 			So(err, ShouldEqual, nil)
-			Convey("Allocation of 10 pages should suceed", func() {
+			Convey("Allocation of 10 pages should succeed", func() {
 				allocated := make([]EdfRange, 10)
 				for i := 0; i < 10; i++ {
 					r, err := mapping.AllocPages(1, 2)
 					So(err, ShouldEqual, nil)
 					allocated[i] = r
 				}
-				Convey("Unmapping the file should suceed", func() {
+				Convey("Unmapping the file should succeed", func() {
 					err = mapping.Unmap(EDF_UNMAP_SYNC)
 					So(err, ShouldEqual, nil)
-					Convey("Remapping the file should suceed", func() {
+					Convey("Remapping the file should succeed", func() {
 						mapping, err = EdfMap(tempFile, EDF_READ_ONLY)
 						Convey("Should get the same allocations back", func() {
 							rr, err := mapping.GetThreadBlocks(2)

base/edf/map_test.go

@@ -8,7 +8,7 @@ import (
 )
 
 func TestAnonMap(t *testing.T) {
-	Convey("Anonymous mapping should suceed", t, func() {
+	Convey("Anonymous mapping should succeed", t, func() {
 		mapping, err := EdfAnonMap()
 		So(err, ShouldEqual, nil)
 		bytes := mapping.m[0]
@@ -39,10 +39,10 @@ func TestFileCreate(t *testing.T) {
 	Convey("Creating a non-existent file should succeed", t, func() {
 		tempFile, err := ioutil.TempFile(os.TempDir(), "TestFileCreate")
 		So(err, ShouldEqual, nil)
-		Convey("Mapping the file should suceed", func() {
+		Convey("Mapping the file should succeed", func() {
 			mapping, err := EdfMap(tempFile, EDF_CREATE)
 			So(err, ShouldEqual, nil)
-			Convey("Unmapping the file should suceed", func() {
+			Convey("Unmapping the file should succeed", func() {
 				err = mapping.Unmap(EDF_UNMAP_SYNC)
 				So(err, ShouldEqual, nil)
 			})
@@ -90,7 +90,7 @@ func TestFileThreadCounter(t *testing.T) {
 	Convey("Creating a non-existent file should succeed", t, func() {
 		tempFile, err := ioutil.TempFile(os.TempDir(), "TestFileCreate")
 		So(err, ShouldEqual, nil)
-		Convey("Mapping the file should suceed", func() {
+		Convey("Mapping the file should succeed", func() {
 			mapping, err := EdfMap(tempFile, EDF_CREATE)
 			So(err, ShouldEqual, nil)
 			Convey("The file should have two threads to start with", func() {

base/edf/thread_test.go

@@ -2,17 +2,17 @@ package edf
 
 import (
 	. "github.com/smartystreets/goconvey/convey"
-	"testing"
 	"os"
+	"testing"
 )
 
-func TestThreadDeserialize(T *testing.T) {
+func TestThreadDeserialize(t *testing.T) {
 	bytes := []byte{0, 0, 0, 6, 83, 89, 83, 84, 69, 77, 0, 0, 0, 1}
-	Convey("Given a byte slice", T, func() {
-		var t Thread
-		size := t.Deserialize(bytes)
+	Convey("Given a byte slice", t, func() {
+		var thread Thread
+		size := thread.Deserialize(bytes)
 		Convey("Decoded name should be SYSTEM", func() {
-			So(t.name, ShouldEqual, "SYSTEM")
+			So(thread.name, ShouldEqual, "SYSTEM")
 		})
 		Convey("Size should be the same as the array", func() {
 			So(size, ShouldEqual, len(bytes))
@@ -20,28 +20,28 @@ func TestThreadDeserialize(T *testing.T) {
 	})
 }
 
-func TestThreadSerialize(T *testing.T) {
-	var t Thread
+func TestThreadSerialize(t *testing.T) {
+	var thread Thread
 	refBytes := []byte{0, 0, 0, 6, 83, 89, 83, 84, 69, 77, 0, 0, 0, 1}
-	t.name = "SYSTEM"
-	t.id = 1
+	thread.name = "SYSTEM"
+	thread.id = 1
 	toBytes := make([]byte, len(refBytes))
-	Convey("Should serialize correctly", T, func() {
-		t.Serialize(toBytes)
+	Convey("Should serialize correctly", t, func() {
+		thread.Serialize(toBytes)
 		So(toBytes, ShouldResemble, refBytes)
 	})
 }
 
-func TestThreadFindAndWrite(T *testing.T) {
-	Convey("Creating a non-existent file should succeed", T, func() {
-		tempFile, err := os.OpenFile("hello.db", os.O_RDWR | os.O_TRUNC | os.O_CREATE, 0700) //ioutil.TempFile(os.TempDir(), "TestFileCreate")
+func TestThreadFindAndWrite(t *testing.T) {
+	Convey("Creating a non-existent file should succeed", t, func() {
+		tempFile, err := os.OpenFile("hello.db", os.O_RDWR|os.O_TRUNC|os.O_CREATE, 0700) //ioutil.TempFile(os.TempDir(), "TestFileCreate")
 		So(err, ShouldEqual, nil)
-		Convey("Mapping the file should suceed", func() {
+		Convey("Mapping the file should succeed", func() {
 			mapping, err := EdfMap(tempFile, EDF_CREATE)
 			So(err, ShouldEqual, nil)
-			Convey("Writing the thread should succeed", func () {
+			Convey("Writing the thread should succeed", func() {
 				t := NewThread(mapping, "MyNameISWhat")
-				Convey("Thread number should be 3", func () {
+				Convey("Thread number should be 3", func() {
 					So(t.id, ShouldEqual, 3)
 				})
 				Convey("Writing the thread should succeed", func() {

base/lazy_sort_test.go

@@ -1,19 +1,18 @@
 package base
 
 import (
-	"fmt"
 	"testing"
 )
 
-func TestLazySortDesc(testEnv *testing.T) {
+func TestLazySortDesc(t *testing.T) {
 	inst1, err := ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 	inst2, err := ParseCSVToInstances("../examples/datasets/iris_sorted_desc.csv", true)
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 
@@ -21,67 +20,67 @@ func TestLazySortDesc(testEnv *testing.T) {
 	as2 := ResolveAllAttributes(inst2)
 
 	if isSortedDesc(inst1, as1[0]) {
-		testEnv.Error("Can't test descending sort order")
+		t.Error("Can't test descending sort order")
 	}
 	if !isSortedDesc(inst2, as2[0]) {
-		testEnv.Error("Reference data not sorted in descending order!")
+		t.Error("Reference data not sorted in descending order!")
 	}
 
 	inst, err := LazySort(inst1, Descending, as1[0:len(as1)-1])
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 	}
 	if !isSortedDesc(inst, as1[0]) {
-		testEnv.Error("Instances are not sorted in descending order")
-		testEnv.Error(inst1)
+		t.Error("Instances are not sorted in descending order")
+		t.Error(inst1)
 	}
 	if !inst2.Equal(inst) {
-		testEnv.Error("Instances don't match")
-		testEnv.Error(inst)
-		testEnv.Error(inst2)
+		t.Error("Instances don't match")
+		t.Error(inst)
+		t.Error(inst2)
 	}
 }
 
-func TestLazySortAsc(testEnv *testing.T) {
+func TestLazySortAsc(t *testing.T) {
 	inst, err := ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	as1 := ResolveAllAttributes(inst)
 	if isSortedAsc(inst, as1[0]) {
-		testEnv.Error("Can't test ascending sort on something ascending already")
+		t.Error("Can't test ascending sort on something ascending already")
 	}
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 	insts, err := LazySort(inst, Ascending, as1)
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 	if !isSortedAsc(insts, as1[0]) {
-		testEnv.Error("Instances are not sorted in ascending order")
-		testEnv.Error(insts)
+		t.Error("Instances are not sorted in ascending order")
+		t.Error(insts)
 	}
 
 	inst2, err := ParseCSVToInstances("../examples/datasets/iris_sorted_asc.csv", true)
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 	as2 := ResolveAllAttributes(inst2)
 	if !isSortedAsc(inst2, as2[0]) {
-		testEnv.Error("This file should be sorted in ascending order")
+		t.Error("This file should be sorted in ascending order")
 	}
 
 	if !inst2.Equal(insts) {
-		testEnv.Error("Instances don't match")
-		testEnv.Error(inst)
-		testEnv.Error(inst2)
+		t.Error("Instances don't match")
+		t.Error(inst)
+		t.Error(inst2)
 	}
 
 	rowStr := insts.RowString(0)
 	ref := "4.30 3.00 1.10 0.10 Iris-setosa"
 	if rowStr != ref {
-		panic(fmt.Sprintf("'%s' != '%s'", rowStr, ref))
+		t.Fatalf("'%s' != '%s'", rowStr, ref)
 	}
 
 }

base/sort_test.go

@@ -32,15 +32,15 @@ func isSortedDesc(inst FixedDataGrid, attr AttributeSpec) bool {
 	return true
 }
 
-func TestSortDesc(testEnv *testing.T) {
+func TestSortDesc(t *testing.T) {
 	inst1, err := ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 	inst2, err := ParseCSVToInstances("../examples/datasets/iris_sorted_desc.csv", true)
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 
@@ -48,57 +48,57 @@ func TestSortDesc(testEnv *testing.T) {
 	as2 := ResolveAllAttributes(inst2)
 
 	if isSortedDesc(inst1, as1[0]) {
-		testEnv.Error("Can't test descending sort order")
+		t.Error("Can't test descending sort order")
 	}
 	if !isSortedDesc(inst2, as2[0]) {
-		testEnv.Error("Reference data not sorted in descending order!")
+		t.Error("Reference data not sorted in descending order!")
 	}
 
 	Sort(inst1, Descending, as1[0:len(as1)-1])
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 	}
 	if !isSortedDesc(inst1, as1[0]) {
-		testEnv.Error("Instances are not sorted in descending order")
-		testEnv.Error(inst1)
+		t.Error("Instances are not sorted in descending order")
+		t.Error(inst1)
 	}
 	if !inst2.Equal(inst1) {
-		testEnv.Error("Instances don't match")
-		testEnv.Error(inst1)
-		testEnv.Error(inst2)
+		t.Error("Instances don't match")
+		t.Error(inst1)
+		t.Error(inst2)
 	}
 }
 
-func TestSortAsc(testEnv *testing.T) {
+func TestSortAsc(t *testing.T) {
 	inst, err := ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	as1 := ResolveAllAttributes(inst)
 	if isSortedAsc(inst, as1[0]) {
-		testEnv.Error("Can't test ascending sort on something ascending already")
+		t.Error("Can't test ascending sort on something ascending already")
 	}
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 	Sort(inst, Ascending, as1[0:1])
 	if !isSortedAsc(inst, as1[0]) {
-		testEnv.Error("Instances are not sorted in ascending order")
-		testEnv.Error(inst)
+		t.Error("Instances are not sorted in ascending order")
+		t.Error(inst)
 	}
 
 	inst2, err := ParseCSVToInstances("../examples/datasets/iris_sorted_asc.csv", true)
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 		return
 	}
 	as2 := ResolveAllAttributes(inst2)
 	if !isSortedAsc(inst2, as2[0]) {
-		testEnv.Error("This file should be sorted in ascending order")
+		t.Error("This file should be sorted in ascending order")
 	}
 
 	if !inst2.Equal(inst) {
-		testEnv.Error("Instances don't match")
-		testEnv.Error(inst)
-		testEnv.Error(inst2)
+		t.Error("Instances don't match")
+		t.Error(inst)
+		t.Error(inst2)
 	}
 
 }

base/util.go

@@ -1,9 +1,6 @@
 package base
 
 import (
-	"bytes"
-	"encoding/binary"
-	"fmt"
 	"math"
 	"unsafe"
 )
@@ -62,29 +59,6 @@ func UnpackBytesToFloat(val []byte) float64 {
 	return *(*float64)(pb)
 }
 
-func xorFloatOp(item float64) float64 {
-	var ret float64
-	var tmp int64
-	buf := bytes.NewBuffer(nil)
-	binary.Write(buf, binary.LittleEndian, item)
-	binary.Read(buf, binary.LittleEndian, &tmp)
-	tmp ^= -1 << 63
-	binary.Write(buf, binary.LittleEndian, tmp)
-	binary.Read(buf, binary.LittleEndian, &ret)
-	return ret
-}
-
-func printFloatByteArr(arr [][]byte) {
-	buf := bytes.NewBuffer(nil)
-	var f float64
-	for _, b := range arr {
-		buf.Write(b)
-		binary.Read(buf, binary.LittleEndian, &f)
-		f = xorFloatOp(f)
-		fmt.Println(f)
-	}
-}
-
 func byteSeqEqual(a, b []byte) bool {
 	if len(a) != len(b) {
 		return false

ensemble/randomforest.go

@@ -2,9 +2,9 @@ package ensemble
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
-	meta "github.com/sjwhitworth/golearn/meta"
-	trees "github.com/sjwhitworth/golearn/trees"
+	"github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/meta"
+	"github.com/sjwhitworth/golearn/trees"
 )
 
 // RandomForest classifies instances using an ensemble
@@ -31,6 +31,15 @@ func NewRandomForest(forestSize int, features int) *RandomForest {
 
 // Fit builds the RandomForest on the specified instances
 func (f *RandomForest) Fit(on base.FixedDataGrid) {
+	numNonClassAttributes := len(base.NonClassAttributes(on))
+	if numNonClassAttributes < f.Features {
+		panic(fmt.Sprintf(
+			"Random forest with %d features cannot fit data grid with %d non-class attributes",
+			f.Features,
+			numNonClassAttributes,
+		))
+	}
+
 	f.Model = new(meta.BaggedModel)
 	f.Model.RandomFeatures = f.Features
 	for i := 0; i < f.ForestSize; i++ {

ensemble/randomforest_test.go

@@ -1,17 +1,16 @@
 package ensemble
 
 import (
-	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	eval "github.com/sjwhitworth/golearn/evaluation"
-	filters "github.com/sjwhitworth/golearn/filters"
+	"github.com/sjwhitworth/golearn/filters"
 	"testing"
 )
 
-func TestRandomForest1(testEnv *testing.T) {
+func TestRandomForest1(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 
 	filt := filters.NewChiMergeFilter(inst, 0.90)
@@ -26,8 +25,6 @@ func TestRandomForest1(testEnv *testing.T) {
 	rf := NewRandomForest(10, 3)
 	rf.Fit(trainData)
 	predictions := rf.Predict(testData)
-	fmt.Println(predictions)
 	confusionMat := eval.GetConfusionMatrix(testData, predictions)
-	fmt.Println(confusionMat)
-	fmt.Println(eval.GetSummary(confusionMat))
+	_ = eval.GetSummary(confusionMat)
 }

evaluation/confusion.go

@@ -3,6 +3,8 @@ package evaluation
 import (
 	"bytes"
 	"fmt"
+	"text/tabwriter"
+
 	"github.com/sjwhitworth/golearn/base"
 )
 
@@ -176,18 +178,22 @@ func GetMacroRecall(c ConfusionMatrix) float64 {
 // ConfusionMatrix
 func GetSummary(c ConfusionMatrix) string {
 	var buffer bytes.Buffer
+	w := new(tabwriter.Writer)
+	w.Init(&buffer, 0, 8, 0, '\t', 0)
+
+	fmt.Fprintln(w, "Reference Class\tTrue Positives\tFalse Positives\tTrue Negatives\tPrecision\tRecall\tF1 Score")
+	fmt.Fprintln(w, "---------------\t--------------\t---------------\t--------------\t---------\t------\t--------")
 	for k := range c {
-		buffer.WriteString(k)
-		buffer.WriteString("\t")
 		tp := GetTruePositives(k, c)
 		fp := GetFalsePositives(k, c)
 		tn := GetTrueNegatives(k, c)
 		prec := GetPrecision(k, c)
 		rec := GetRecall(k, c)
 		f1 := GetF1Score(k, c)
-		buffer.WriteString(fmt.Sprintf("%.0f\t%.0f\t%.0f\t%.4f\t%.4f\t%.4f\n", tp, fp, tn, prec, rec, f1))
-	}
 
+		fmt.Fprintf(w, "%s\t%.0f\t%.0f\t%.0f\t%.4f\t%.4f\t%.4f\n", k, tp, fp, tn, prec, rec, f1)
+	}
+	w.Flush()
 	buffer.WriteString(fmt.Sprintf("Overall accuracy: %.4f\n", GetAccuracy(c)))
 
 	return buffer.String()

evaluation/confusion_test.go

@@ -5,7 +5,7 @@ import (
 	"testing"
 )
 
-func TestMetrics(testEnv *testing.T) {
+func TestMetrics(t *testing.T) {
 	confusionMat := make(ConfusionMatrix)
 	confusionMat["a"] = make(map[string]int)
 	confusionMat["b"] = make(map[string]int)
@@ -16,89 +16,89 @@ func TestMetrics(testEnv *testing.T) {
 
 	tp := GetTruePositives("a", confusionMat)
 	if math.Abs(tp-75) >= 1 {
-		testEnv.Error(tp)
+		t.Error(tp)
 	}
 	tp = GetTruePositives("b", confusionMat)
 	if math.Abs(tp-10) >= 1 {
-		testEnv.Error(tp)
+		t.Error(tp)
 	}
 
 	fn := GetFalseNegatives("a", confusionMat)
 	if math.Abs(fn-5) >= 1 {
-		testEnv.Error(fn)
+		t.Error(fn)
 	}
 	fn = GetFalseNegatives("b", confusionMat)
 	if math.Abs(fn-10) >= 1 {
-		testEnv.Error(fn)
+		t.Error(fn)
 	}
 
 	tn := GetTrueNegatives("a", confusionMat)
 	if math.Abs(tn-10) >= 1 {
-		testEnv.Error(tn)
+		t.Error(tn)
 	}
 	tn = GetTrueNegatives("b", confusionMat)
 	if math.Abs(tn-75) >= 1 {
-		testEnv.Error(tn)
+		t.Error(tn)
 	}
 
 	fp := GetFalsePositives("a", confusionMat)
 	if math.Abs(fp-10) >= 1 {
-		testEnv.Error(fp)
+		t.Error(fp)
 	}
 
 	fp = GetFalsePositives("b", confusionMat)
 	if math.Abs(fp-5) >= 1 {
-		testEnv.Error(fp)
+		t.Error(fp)
 	}
 
 	precision := GetPrecision("a", confusionMat)
 	recall := GetRecall("a", confusionMat)
 
 	if math.Abs(precision-0.88) >= 0.01 {
-		testEnv.Error(precision)
+		t.Error(precision)
 	}
 
 	if math.Abs(recall-0.94) >= 0.01 {
-		testEnv.Error(recall)
+		t.Error(recall)
 	}
 
 	precision = GetPrecision("b", confusionMat)
 	recall = GetRecall("b", confusionMat)
 	if math.Abs(precision-0.666) >= 0.01 {
-		testEnv.Error(precision)
+		t.Error(precision)
 	}
 
 	if math.Abs(recall-0.50) >= 0.01 {
-		testEnv.Error(recall)
+		t.Error(recall)
 	}
 
 	precision = GetMicroPrecision(confusionMat)
 	if math.Abs(precision-0.85) >= 0.01 {
-		testEnv.Error(precision)
+		t.Error(precision)
 	}
 
 	recall = GetMicroRecall(confusionMat)
 	if math.Abs(recall-0.85) >= 0.01 {
-		testEnv.Error(recall)
+		t.Error(recall)
 	}
 
 	precision = GetMacroPrecision(confusionMat)
 	if math.Abs(precision-0.775) >= 0.01 {
-		testEnv.Error(precision)
+		t.Error(precision)
 	}
 
 	recall = GetMacroRecall(confusionMat)
 	if math.Abs(recall-0.719) > 0.01 {
-		testEnv.Error(recall)
+		t.Error(recall)
 	}
 
 	fmeasure := GetF1Score("a", confusionMat)
 	if math.Abs(fmeasure-0.91) >= 0.1 {
-		testEnv.Error(fmeasure)
+		t.Error(fmeasure)
 	}
 
 	accuracy := GetAccuracy(confusionMat)
 	if math.Abs(accuracy-0.85) >= 0.1 {
-		testEnv.Error(accuracy)
+		t.Error(accuracy)
 	}
 }

examples/instances/instances.go

@@ -4,7 +4,7 @@ package main
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 )
 
 func main() {

examples/knnclassifier/knnclassifier_iris.go

@@ -2,9 +2,9 @@ package main
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
-	evaluation "github.com/sjwhitworth/golearn/evaluation"
-	knn "github.com/sjwhitworth/golearn/knn"
+	"github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/evaluation"
+	"github.com/sjwhitworth/golearn/knn"
 )
 
 func main() {

examples/trees/trees.go

@@ -4,11 +4,11 @@ package main
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
-	ensemble "github.com/sjwhitworth/golearn/ensemble"
+	"github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/ensemble"
 	eval "github.com/sjwhitworth/golearn/evaluation"
-	filters "github.com/sjwhitworth/golearn/filters"
-	trees "github.com/sjwhitworth/golearn/trees"
+	"github.com/sjwhitworth/golearn/filters"
+	"github.com/sjwhitworth/golearn/trees"
 	"math/rand"
 	"time"
 )

ext/make.go

@@ -45,7 +45,7 @@ func main() {
 		return
 	}
 
-	os.Mkdir("lib", os.ModeDir | 0777)
+	os.Mkdir("lib", os.ModeDir|0777)
 
 	log.Println("Installing libs")
 	if runtime.GOOS == "windows" {

filters/binary_test.go

@@ -1,7 +1,6 @@
 package filters
 
 import (
-	"fmt"
 	"github.com/sjwhitworth/golearn/base"
 	. "github.com/smartystreets/goconvey/convey"
 	"testing"
@@ -38,9 +37,6 @@ func TestBinaryFilterClassPreservation(t *testing.T) {
 			So(attrMap["arbitraryClass_there"], ShouldEqual, true)
 			So(attrMap["arbitraryClass_world"], ShouldEqual, true)
 		})
-
-		fmt.Println(instF)
-
 	})
 }
 
@@ -91,7 +87,7 @@ func TestBinaryFilter(t *testing.T) {
 				name := a.GetName()
 				_, ok := origMap[name]
 				if !ok {
-					t.Error(fmt.Sprintf("Weird: %s", name))
+					t.Errorf("Weird: %s", name)
 				}
 				origMap[name] = true
 			}

filters/binning.go

@@ -2,7 +2,7 @@ package filters
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	"math"
 )
 

filters/binning_test.go

@@ -1,7 +1,7 @@
 package filters
 
 import (
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	. "github.com/smartystreets/goconvey/convey"
 	"testing"
 )
@@ -11,12 +11,12 @@ func TestBinning(t *testing.T) {
 		// Read the data
 		inst1, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 		if err != nil {
-			panic(err)
+			t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 		}
 
 		inst2, err := base.ParseCSVToInstances("../examples/datasets/iris_binned.csv", true)
 		if err != nil {
-			panic(err)
+			t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 		}
 		//
 		// Construct the binning filter

filters/chimerge.go

@@ -2,7 +2,7 @@ package filters
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	"math"
 )
 

filters/chimerge_funcs.go

@@ -2,7 +2,6 @@ package filters
 
 import (
 	"github.com/sjwhitworth/golearn/base"
-	"fmt"
 	"math"
 )
 
@@ -185,21 +184,3 @@ func chiMergeMergeZipAdjacent(freq []*FrequencyTableEntry, minIndex int) []*Freq
 	freq = append(lowerSlice, upperSlice...)
 	return freq
 }
-
-func chiMergePrintTable(freq []*FrequencyTableEntry) {
-	classes := chiCountClasses(freq)
-	fmt.Printf("Attribute value\t")
-	for k := range classes {
-		fmt.Printf("\t%s", k)
-	}
-	fmt.Printf("\tTotal\n")
-	for _, f := range freq {
-		fmt.Printf("%.2f\t", f.Value)
-		total := 0
-		for k := range classes {
-			fmt.Printf("\t%d", f.Frequency[k])
-			total += f.Frequency[k]
-		}
-		fmt.Printf("\t%d\n", total)
-	}
-}

filters/chimerge_test.go

@@ -1,113 +1,109 @@
 package filters
 
 import (
-	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	"math"
 	"testing"
 )
 
-func TestChiMFreqTable(testEnv *testing.T) {
-
+func TestChiMFreqTable(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/chim.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 
 	freq := ChiMBuildFrequencyTable(inst.AllAttributes()[0], inst)
 
 	if freq[0].Frequency["c1"] != 1 {
-		testEnv.Error("Wrong frequency")
+		t.Error("Wrong frequency")
 	}
 	if freq[0].Frequency["c3"] != 4 {
-		testEnv.Errorf("Wrong frequency %s", freq[1])
+		t.Errorf("Wrong frequency %s", freq[1])
 	}
 	if freq[10].Frequency["c2"] != 1 {
-		testEnv.Error("Wrong frequency")
+		t.Error("Wrong frequency")
 	}
 }
 
-func TestChiClassCounter(testEnv *testing.T) {
+func TestChiClassCounter(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/chim.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 	freq := ChiMBuildFrequencyTable(inst.AllAttributes()[0], inst)
 	classes := chiCountClasses(freq)
 	if classes["c1"] != 27 {
-		testEnv.Error(classes)
+		t.Error(classes)
 	}
 	if classes["c2"] != 12 {
-		testEnv.Error(classes)
+		t.Error(classes)
 	}
 	if classes["c3"] != 21 {
-		testEnv.Error(classes)
+		t.Error(classes)
 	}
 }
 
-func TestStatisticValues(testEnv *testing.T) {
+func TestStatisticValues(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/chim.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 	freq := ChiMBuildFrequencyTable(inst.AllAttributes()[0], inst)
 	chiVal := chiComputeStatistic(freq[5], freq[6])
 	if math.Abs(chiVal-1.89) > 0.01 {
-		testEnv.Error(chiVal)
+		t.Error(chiVal)
 	}
 
 	chiVal = chiComputeStatistic(freq[1], freq[2])
 	if math.Abs(chiVal-1.08) > 0.01 {
-		testEnv.Error(chiVal)
+		t.Error(chiVal)
 	}
 }
 
-func TestChiSquareDistValues(testEnv *testing.T) {
+func TestChiSquareDistValues(t *testing.T) {
 	chiVal1 := chiSquaredPercentile(2, 4.61)
 	chiVal2 := chiSquaredPercentile(3, 7.82)
 	chiVal3 := chiSquaredPercentile(4, 13.28)
 	if math.Abs(chiVal1-0.90) > 0.001 {
-		testEnv.Error(chiVal1)
+		t.Error(chiVal1)
 	}
 	if math.Abs(chiVal2-0.95) > 0.001 {
-		testEnv.Error(chiVal2)
+		t.Error(chiVal2)
 	}
 	if math.Abs(chiVal3-0.99) > 0.001 {
-		testEnv.Error(chiVal3)
+		t.Error(chiVal3)
 	}
 }
 
-func TestChiMerge1(testEnv *testing.T) {
-
-	// Read the data
+func TestChiMerge1(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/chim.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 	_, rows := inst.Size()
 
 	freq := chiMerge(inst, inst.AllAttributes()[0], 0.90, 0, rows)
 	if len(freq) != 3 {
-		testEnv.Error("Wrong length")
+		t.Error("Wrong length")
 	}
 	if freq[0].Value != 1.3 {
-		testEnv.Error(freq[0])
+		t.Error(freq[0])
 	}
 	if freq[1].Value != 56.2 {
-		testEnv.Error(freq[1])
+		t.Error(freq[1])
 	}
 	if freq[2].Value != 87.1 {
-		testEnv.Error(freq[2])
+		t.Error(freq[2])
 	}
 }
 
-func TestChiMerge2(testEnv *testing.T) {
+func TestChiMerge2(t *testing.T) {
 	//
 	// See http://sci2s.ugr.es/keel/pdf/algorithm/congreso/1992-Kerber-ChimErge-AAAI92.pdf
 	//   Randy Kerber, ChiMerge: Discretisation of Numeric Attributes, 1992
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 
 	// Sort the instances
@@ -115,35 +111,35 @@ func TestChiMerge2(testEnv *testing.T) {
 	sortAttrSpecs := base.ResolveAttributes(inst, allAttrs)[0:1]
 	instSorted, err := base.Sort(inst, base.Ascending, sortAttrSpecs)
 	if err != nil {
-		panic(err)
+		t.Fatalf("Sort failed: %s", err.Error())
 	}
 
 	// Perform Chi-Merge
 	_, rows := inst.Size()
 	freq := chiMerge(instSorted, allAttrs[0], 0.90, 0, rows)
 	if len(freq) != 5 {
-		testEnv.Errorf("Wrong length (%d)", len(freq))
-		testEnv.Error(freq)
+		t.Errorf("Wrong length (%d)", len(freq))
+		t.Error(freq)
 	}
 	if freq[0].Value != 4.3 {
-		testEnv.Error(freq[0])
+		t.Error(freq[0])
 	}
 	if freq[1].Value != 5.5 {
-		testEnv.Error(freq[1])
+		t.Error(freq[1])
 	}
 	if freq[2].Value != 5.8 {
-		testEnv.Error(freq[2])
+		t.Error(freq[2])
 	}
 	if freq[3].Value != 6.3 {
-		testEnv.Error(freq[3])
+		t.Error(freq[3])
 	}
 	if freq[4].Value != 7.1 {
-		testEnv.Error(freq[4])
+		t.Error(freq[4])
 	}
 }
 
 /*
-func TestChiMerge3(testEnv *testing.T) {
+func TestChiMerge3(t *testing.T) {
 	// See http://sci2s.ugr.es/keel/pdf/algorithm/congreso/1992-Kerber-ChimErge-AAAI92.pdf
 	//   Randy Kerber, ChiMerge: Discretisation of Numeric Attributes, 1992
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
@@ -153,7 +149,7 @@ func TestChiMerge3(testEnv *testing.T) {
 
 	insts, err := base.LazySort(inst, base.Ascending, base.ResolveAllAttributes(inst, inst.AllAttributes()))
 	if err != nil {
-		testEnv.Error(err)
+		t.Error(err)
 	}
 	filt := NewChiMergeFilter(inst, 0.90)
 	filt.AddAttribute(inst.AllAttributes()[0])
@@ -176,12 +172,12 @@ func TestChiMerge3(testEnv *testing.T) {
 }
 */
 
-func TestChiMerge4(testEnv *testing.T) {
+func TestChiMerge4(t *testing.T) {
 	// See http://sci2s.ugr.es/keel/pdf/algorithm/congreso/1992-Kerber-ChimErge-AAAI92.pdf
 	//   Randy Kerber, ChiMerge: Discretisation of Numeric Attributes, 1992
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 
 	filt := NewChiMergeFilter(inst, 0.90)
@@ -189,13 +185,13 @@ func TestChiMerge4(testEnv *testing.T) {
 	filt.AddAttribute(inst.AllAttributes()[1])
 	filt.Train()
 	instf := base.NewLazilyFilteredInstances(inst, filt)
-	fmt.Println(instf)
-	fmt.Println(instf.String())
 	clsAttrs := instf.AllClassAttributes()
 	if len(clsAttrs) != 1 {
-		panic(fmt.Sprintf("%d != %d", len(clsAttrs), 1))
+		t.Fatalf("%d != %d", len(clsAttrs), 1)
 	}
-	if clsAttrs[0].GetName() != "Species" {
-		panic("Class Attribute wrong!")
+	firstClassAttributeName := clsAttrs[0].GetName()
+	expectedClassAttributeName := "Species"
+	if firstClassAttributeName != expectedClassAttributeName {
+		t.Fatalf("Expected class attribute '%s'; actual class attribute '%s'", expectedClassAttributeName, firstClassAttributeName)
 	}
 }

filters/disc.go

@@ -2,7 +2,7 @@ package filters
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 )
 
 type AbstractDiscretizeFilter struct {

filters/float_test.go

@@ -1,7 +1,6 @@
 package filters
 
 import (
-	"fmt"
 	"github.com/sjwhitworth/golearn/base"
 	. "github.com/smartystreets/goconvey/convey"
 	"testing"
@@ -54,7 +53,7 @@ func TestFloatFilter(t *testing.T) {
 				name := a.GetName()
 				_, ok := origMap[name]
 				if !ok {
-					t.Error(fmt.Sprintf("Weird: %s", name))
+					t.Errorf("Weird: %s", name)
 				}
 				origMap[name] = true
 			}

knn/knn.go

@@ -5,7 +5,7 @@ package knn
 
 import (
 	"github.com/gonum/matrix/mat64"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	pairwiseMetrics "github.com/sjwhitworth/golearn/metrics/pairwise"
 	util "github.com/sjwhitworth/golearn/utilities"
 )

linear_models/linear_regression_test.go

@@ -1,7 +1,6 @@
 package linear_models
 
 import (
-	"fmt"
 	"testing"
 
 	"github.com/sjwhitworth/golearn/base"
@@ -54,11 +53,7 @@ func TestLinearRegression(t *testing.T) {
 		t.Fatal(err)
 	}
 
-	_, rows := predictions.Size()
-
-	for i := 0; i < rows; i++ {
-		fmt.Printf("Expected: %s || Predicted: %s\n", base.GetClass(testData, i), base.GetClass(predictions, i))
-	}
+	_, _ = predictions.Size()
 }
 
 func BenchmarkLinearRegressionOneRow(b *testing.B) {

linear_models/logistic.go

@@ -2,7 +2,7 @@ package linear_models
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 )
 
 type LogisticRegression struct {

meta/bagging.go

@@ -2,7 +2,7 @@ package meta
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	"math/rand"
 	"runtime"
 	"strings"

meta/bagging_test.go

@@ -1,20 +1,19 @@
 package meta
 
 import (
-	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	eval "github.com/sjwhitworth/golearn/evaluation"
-	filters "github.com/sjwhitworth/golearn/filters"
-	trees "github.com/sjwhitworth/golearn/trees"
+	"github.com/sjwhitworth/golearn/filters"
+	"github.com/sjwhitworth/golearn/trees"
 	"math/rand"
 	"testing"
 	"time"
 )
 
-func BenchmarkBaggingRandomForestFit(testEnv *testing.B) {
+func BenchmarkBaggingRandomForestFit(t *testing.B) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 
 	rand.Seed(time.Now().UnixNano())
@@ -24,20 +23,22 @@ func BenchmarkBaggingRandomForestFit(testEnv *testing.B) {
 	}
 	filt.Train()
 	instf := base.NewLazilyFilteredInstances(inst, filt)
+
 	rf := new(BaggedModel)
 	for i := 0; i < 10; i++ {
 		rf.AddModel(trees.NewRandomTree(2))
 	}
-	testEnv.ResetTimer()
+
+	t.ResetTimer()
 	for i := 0; i < 20; i++ {
 		rf.Fit(instf)
 	}
 }
 
-func BenchmarkBaggingRandomForestPredict(testEnv *testing.B) {
+func BenchmarkBaggingRandomForestPredict(t *testing.B) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 
 	rand.Seed(time.Now().UnixNano())
@@ -47,25 +48,27 @@ func BenchmarkBaggingRandomForestPredict(testEnv *testing.B) {
 	}
 	filt.Train()
 	instf := base.NewLazilyFilteredInstances(inst, filt)
+
 	rf := new(BaggedModel)
 	for i := 0; i < 10; i++ {
 		rf.AddModel(trees.NewRandomTree(2))
 	}
+
 	rf.Fit(instf)
-	testEnv.ResetTimer()
+	t.ResetTimer()
 	for i := 0; i < 20; i++ {
 		rf.Predict(instf)
 	}
 }
 
-func TestRandomForest1(testEnv *testing.T) {
+func TestRandomForest1(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
+	trainData, testData := base.InstancesTrainTestSplit(inst, 0.6)
 
 	rand.Seed(time.Now().UnixNano())
-	trainData, testData := base.InstancesTrainTestSplit(inst, 0.6)
 	filt := filters.NewChiMergeFilter(inst, 0.90)
 	for _, a := range base.NonClassFloatAttributes(inst) {
 		filt.AddAttribute(a)
@@ -73,17 +76,14 @@ func TestRandomForest1(testEnv *testing.T) {
 	filt.Train()
 	trainDataf := base.NewLazilyFilteredInstances(trainData, filt)
 	testDataf := base.NewLazilyFilteredInstances(testData, filt)
+
 	rf := new(BaggedModel)
 	for i := 0; i < 10; i++ {
 		rf.AddModel(trees.NewRandomTree(2))
 	}
+
 	rf.Fit(trainDataf)
-	fmt.Println(rf)
 	predictions := rf.Predict(testDataf)
-	fmt.Println(predictions)
 	confusionMat := eval.GetConfusionMatrix(testDataf, predictions)
-	fmt.Println(confusionMat)
-	fmt.Println(eval.GetMacroPrecision(confusionMat))
-	fmt.Println(eval.GetMacroRecall(confusionMat))
-	fmt.Println(eval.GetSummary(confusionMat))
+	_ = eval.GetSummary(confusionMat)
 }

naive/bernoulli_nb.go

@@ -2,7 +2,7 @@ package naive
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	"math"
 )
 

neural/layered_test.go

@@ -1,7 +1,6 @@
 package neural
 
 import (
-	"fmt"
 	"github.com/gonum/matrix/mat64"
 	"github.com/sjwhitworth/golearn/base"
 	. "github.com/smartystreets/goconvey/convey"
@@ -13,7 +12,6 @@ func TestLayerStructureNoHidden(t *testing.T) {
 	Convey("Creating a network...", t, func() {
 		XORData, err := base.ParseCSVToInstances("xor.csv", false)
 		So(err, ShouldEqual, nil)
-		fmt.Println(XORData)
 		Convey("Create a MultiLayerNet with no layers...", func() {
 			net := NewMultiLayerNet(make([]int, 0))
 			net.MaxIterations = 0
@@ -73,8 +71,6 @@ func TestLayerStructureNoHidden(t *testing.T) {
 			})
 
 			Convey("The right nodes should be connected in the network...", func() {
-
-				fmt.Println(net.network)
 				So(net.network.GetWeight(1, 1), ShouldAlmostEqual, 1.000)
 				So(net.network.GetWeight(2, 2), ShouldAlmostEqual, 1.000)
 				So(net.network.GetWeight(1, 3), ShouldNotAlmostEqual, 0.000)
@@ -118,7 +114,6 @@ func TestLayeredXOR(t *testing.T) {
 		XORData, err := base.ParseCSVToInstances("xor.csv", false)
 		So(err, ShouldEqual, nil)
 
-		fmt.Println(XORData)
 		net := NewMultiLayerNet([]int{3})
 		net.MaxIterations = 20000
 		net.Fit(XORData)
@@ -126,8 +121,6 @@ func TestLayeredXOR(t *testing.T) {
 		Convey("After running for 20000 iterations, should have some predictive power...", func() {
 
 			Convey("The right nodes should be connected in the network...", func() {
-
-				fmt.Println(net.network)
 				So(net.network.GetWeight(1, 1), ShouldAlmostEqual, 1.000)
 				So(net.network.GetWeight(2, 2), ShouldAlmostEqual, 1.000)
 
@@ -138,7 +131,6 @@ func TestLayeredXOR(t *testing.T) {
 			})
 			out := mat64.NewDense(6, 1, []float64{1.0, 0.0, 0.0, 0.0, 0.0, 0.0})
 			net.network.Activate(out, 2)
-			fmt.Println(out)
 			So(out.At(5, 0), ShouldAlmostEqual, 1.0, 0.1)
 
 			Convey("And Predict() should do OK too...", func() {
@@ -148,7 +140,7 @@ func TestLayeredXOR(t *testing.T) {
 				for _, a := range pred.AllAttributes() {
 					af, ok := a.(*base.FloatAttribute)
 					if !ok {
-						panic("All of these should be FloatAttributes!")
+						t.Fatalf("Expected all attributes to be FloatAttributes; actually some were not")
 					}
 					af.Precision = 1
 				}

neural/network_test.go

@@ -1,7 +1,6 @@
 package neural
 
 import (
-	"fmt"
 	"github.com/gonum/matrix/mat64"
 	. "github.com/smartystreets/goconvey/convey"
 	"testing"
@@ -61,7 +60,6 @@ func TestNetworkWith1Layer(t *testing.T) {
 					for i := 1; i <= 6; i++ {
 						for j := 1; j <= 6; j++ {
 							v := n.GetWeight(i, j)
-							fmt.Println(i, j, v)
 							switch i {
 							case 1:
 								switch j {

trees/entropy.go

@@ -1,7 +1,7 @@
 package trees
 
 import (
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	"math"
 )
 

trees/id3.go

@@ -3,7 +3,7 @@ package trees
 import (
 	"bytes"
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	eval "github.com/sjwhitworth/golearn/evaluation"
 	"sort"
 )

trees/random.go

@@ -2,7 +2,7 @@ package trees
 
 import (
 	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	"math/rand"
 )
 

trees/tree_test.go

@@ -1,19 +1,19 @@
 package trees
 
 import (
-	"fmt"
-	base "github.com/sjwhitworth/golearn/base"
+	"github.com/sjwhitworth/golearn/base"
 	eval "github.com/sjwhitworth/golearn/evaluation"
-	filters "github.com/sjwhitworth/golearn/filters"
+	"github.com/sjwhitworth/golearn/filters"
 	"math"
 	"testing"
 )
 
-func TestRandomTree(testEnv *testing.T) {
+func TestRandomTree(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
+
 	filt := filters.NewChiMergeFilter(inst, 0.90)
 	for _, a := range base.NonClassFloatAttributes(inst) {
 		filt.AddAttribute(a)
@@ -23,17 +23,17 @@ func TestRandomTree(testEnv *testing.T) {
 
 	r := new(RandomTreeRuleGenerator)
 	r.Attributes = 2
-	fmt.Println(instf)
-	root := InferID3Tree(instf, r)
-	fmt.Println(root)
+
+	_ = InferID3Tree(instf, r)
 }
 
-func TestRandomTreeClassification(testEnv *testing.T) {
+func TestRandomTreeClassification(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 	trainData, testData := base.InstancesTrainTestSplit(inst, 0.6)
+
 	filt := filters.NewChiMergeFilter(inst, 0.90)
 	for _, a := range base.NonClassFloatAttributes(inst) {
 		filt.AddAttribute(a)
@@ -44,23 +44,21 @@ func TestRandomTreeClassification(testEnv *testing.T) {
 
 	r := new(RandomTreeRuleGenerator)
 	r.Attributes = 2
+
 	root := InferID3Tree(trainDataF, r)
-	fmt.Println(root)
+
 	predictions := root.Predict(testDataF)
-	fmt.Println(predictions)
 	confusionMat := eval.GetConfusionMatrix(testDataF, predictions)
-	fmt.Println(confusionMat)
-	fmt.Println(eval.GetMacroPrecision(confusionMat))
-	fmt.Println(eval.GetMacroRecall(confusionMat))
-	fmt.Println(eval.GetSummary(confusionMat))
+	_ = eval.GetSummary(confusionMat)
 }
 
-func TestRandomTreeClassification2(testEnv *testing.T) {
+func TestRandomTreeClassification2(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 	trainData, testData := base.InstancesTrainTestSplit(inst, 0.4)
+
 	filt := filters.NewChiMergeFilter(inst, 0.90)
 	for _, a := range base.NonClassFloatAttributes(inst) {
 		filt.AddAttribute(a)
@@ -71,22 +69,19 @@ func TestRandomTreeClassification2(testEnv *testing.T) {
 
 	root := NewRandomTree(2)
 	root.Fit(trainDataF)
-	fmt.Println(root)
+
 	predictions := root.Predict(testDataF)
-	fmt.Println(predictions)
 	confusionMat := eval.GetConfusionMatrix(testDataF, predictions)
-	fmt.Println(confusionMat)
-	fmt.Println(eval.GetMacroPrecision(confusionMat))
-	fmt.Println(eval.GetMacroRecall(confusionMat))
-	fmt.Println(eval.GetSummary(confusionMat))
+	_ = eval.GetSummary(confusionMat)
 }
 
-func TestPruning(testEnv *testing.T) {
+func TestPruning(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 	trainData, testData := base.InstancesTrainTestSplit(inst, 0.6)
+
 	filt := filters.NewChiMergeFilter(inst, 0.90)
 	for _, a := range base.NonClassFloatAttributes(inst) {
 		filt.AddAttribute(a)
@@ -99,17 +94,13 @@ func TestPruning(testEnv *testing.T) {
 	fittrainData, fittestData := base.InstancesTrainTestSplit(trainDataF, 0.6)
 	root.Fit(fittrainData)
 	root.Prune(fittestData)
-	fmt.Println(root)
+
 	predictions := root.Predict(testDataF)
-	fmt.Println(predictions)
 	confusionMat := eval.GetConfusionMatrix(testDataF, predictions)
-	fmt.Println(confusionMat)
-	fmt.Println(eval.GetMacroPrecision(confusionMat))
-	fmt.Println(eval.GetMacroRecall(confusionMat))
-	fmt.Println(eval.GetSummary(confusionMat))
+	_ = eval.GetSummary(confusionMat)
 }
 
-func TestInformationGain(testEnv *testing.T) {
+func TestInformationGain(t *testing.T) {
 	outlook := make(map[string]map[string]int)
 	outlook["sunny"] = make(map[string]int)
 	outlook["overcast"] = make(map[string]int)
@@ -122,16 +113,14 @@ func TestInformationGain(testEnv *testing.T) {
 
 	entropy := getSplitEntropy(outlook)
 	if math.Abs(entropy-0.694) > 0.001 {
-		testEnv.Error(entropy)
+		t.Error(entropy)
 	}
 }
 
-func TestID3Inference(testEnv *testing.T) {
-
-	// Import the "PlayTennis" dataset
+func TestID3Inference(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/tennis.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 
 	// Build the decision tree
@@ -141,81 +130,71 @@ func TestID3Inference(testEnv *testing.T) {
 	// Verify the tree
 	// First attribute should be "outlook"
 	if root.SplitAttr.GetName() != "outlook" {
-		testEnv.Error(root)
+		t.Error(root)
 	}
 	sunnyChild := root.Children["sunny"]
 	overcastChild := root.Children["overcast"]
 	rainyChild := root.Children["rainy"]
 	if sunnyChild.SplitAttr.GetName() != "humidity" {
-		testEnv.Error(sunnyChild)
+		t.Error(sunnyChild)
 	}
 	if rainyChild.SplitAttr.GetName() != "windy" {
-		fmt.Println(rainyChild.SplitAttr)
-		testEnv.Error(rainyChild)
+		t.Error(rainyChild)
 	}
 	if overcastChild.SplitAttr != nil {
-		testEnv.Error(overcastChild)
+		t.Error(overcastChild)
 	}
 
 	sunnyLeafHigh := sunnyChild.Children["high"]
 	sunnyLeafNormal := sunnyChild.Children["normal"]
 	if sunnyLeafHigh.Class != "no" {
-		testEnv.Error(sunnyLeafHigh)
+		t.Error(sunnyLeafHigh)
 	}
 	if sunnyLeafNormal.Class != "yes" {
-		testEnv.Error(sunnyLeafNormal)
+		t.Error(sunnyLeafNormal)
 	}
 	windyLeafFalse := rainyChild.Children["false"]
 	windyLeafTrue := rainyChild.Children["true"]
 	if windyLeafFalse.Class != "yes" {
-		testEnv.Error(windyLeafFalse)
+		t.Error(windyLeafFalse)
 	}
 	if windyLeafTrue.Class != "no" {
-		testEnv.Error(windyLeafTrue)
+		t.Error(windyLeafTrue)
 	}
 
 	if overcastChild.Class != "yes" {
-		testEnv.Error(overcastChild)
+		t.Error(overcastChild)
 	}
 }
 
-func TestID3Classification(testEnv *testing.T) {
+func TestID3Classification(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
-	fmt.Println(inst)
+
 	filt := filters.NewBinningFilter(inst, 10)
 	for _, a := range base.NonClassFloatAttributes(inst) {
 		filt.AddAttribute(a)
 	}
 	filt.Train()
-	fmt.Println(filt)
 	instf := base.NewLazilyFilteredInstances(inst, filt)
-	fmt.Println("INSTFA", instf.AllAttributes())
-	fmt.Println("INSTF", instf)
+
 	trainData, testData := base.InstancesTrainTestSplit(instf, 0.70)
 
 	// Build the decision tree
 	rule := new(InformationGainRuleGenerator)
 	root := InferID3Tree(trainData, rule)
-	fmt.Println(root)
+
 	predictions := root.Predict(testData)
-	fmt.Println(predictions)
 	confusionMat := eval.GetConfusionMatrix(testData, predictions)
-	fmt.Println(confusionMat)
-	fmt.Println(eval.GetMacroPrecision(confusionMat))
-	fmt.Println(eval.GetMacroRecall(confusionMat))
-	fmt.Println(eval.GetSummary(confusionMat))
+	_ = eval.GetSummary(confusionMat)
 }
 
-func TestID3(testEnv *testing.T) {
-
-	// Import the "PlayTennis" dataset
+func TestID3(t *testing.T) {
 	inst, err := base.ParseCSVToInstances("../examples/datasets/tennis.csv", true)
-	fmt.Println(inst)
 	if err != nil {
-		panic(err)
+		t.Fatal("Unable to parse CSV to instances: %s", err.Error())
 	}
 
 	// Build the decision tree
@@ -226,40 +205,40 @@ func TestID3(testEnv *testing.T) {
 	// Verify the tree
 	// First attribute should be "outlook"
 	if root.SplitAttr.GetName() != "outlook" {
-		testEnv.Error(root)
+		t.Error(root)
 	}
 	sunnyChild := root.Children["sunny"]
 	overcastChild := root.Children["overcast"]
 	rainyChild := root.Children["rainy"]
 	if sunnyChild.SplitAttr.GetName() != "humidity" {
-		testEnv.Error(sunnyChild)
+		t.Error(sunnyChild)
 	}
 	if rainyChild.SplitAttr.GetName() != "windy" {
-		testEnv.Error(rainyChild)
+		t.Error(rainyChild)
 	}
 	if overcastChild.SplitAttr != nil {
-		testEnv.Error(overcastChild)
+		t.Error(overcastChild)
 	}
 
 	sunnyLeafHigh := sunnyChild.Children["high"]
 	sunnyLeafNormal := sunnyChild.Children["normal"]
 	if sunnyLeafHigh.Class != "no" {
-		testEnv.Error(sunnyLeafHigh)
+		t.Error(sunnyLeafHigh)
 	}
 	if sunnyLeafNormal.Class != "yes" {
-		testEnv.Error(sunnyLeafNormal)
+		t.Error(sunnyLeafNormal)
 	}
 
 	windyLeafFalse := rainyChild.Children["false"]
 	windyLeafTrue := rainyChild.Children["true"]
 	if windyLeafFalse.Class != "yes" {
-		testEnv.Error(windyLeafFalse)
+		t.Error(windyLeafFalse)
 	}
 	if windyLeafTrue.Class != "no" {
-		testEnv.Error(windyLeafTrue)
+		t.Error(windyLeafTrue)
 	}
 
 	if overcastChild.Class != "yes" {
-		testEnv.Error(overcastChild)
+		t.Error(overcastChild)
 	}
 }