Merge new changes

.gitignore

@@ -14,3 +14,6 @@
 
 # go test coverprofiles
 *.coverprofile
+
+#vim 
+*.sw*

.travis.yml

@@ -1,10 +1,9 @@
 language: go
 go:
- - "1.8"
- - "1.9"
- - "1.10"
+ - 1.13.x
+ - 1.14.x
 env:
- # Temporary workaround for go 1.6
+ # Temporary workaround for Go 1.6+
  - GODEBUG=cgocheck=0
 before_install:
  - sudo apt-get update -qq
@@ -14,3 +13,9 @@ before_install:
 install:
  - go get github.com/smartystreets/goconvey/convey
  - go get -v ./...
+
+script:
+ - ./coverage.sh
+
+after_success:
+ - bash <(curl -s https://codecov.io/bash)

README.md

@@ -4,6 +4,7 @@ GoLearn
 <img src="http://talks.golang.org/2013/advconc/gopherhat.jpg" width=125><br>
 [![GoDoc](https://godoc.org/github.com/sjwhitworth/golearn?status.png)](https://godoc.org/github.com/sjwhitworth/golearn)
 [![Build Status](https://travis-ci.org/sjwhitworth/golearn.png?branch=master)](https://travis-ci.org/sjwhitworth/golearn)<br>
+[![Code Coverage](https://codecov.io/gh/sjwhitworth/golearn/branch/master/graph/badge.svg)](https://codecov.io/gh/sjwhitworth/golearn)
 
 [![Support via Gittip](https://rawgithub.com/twolfson/gittip-badge/0.2.0/dist/gittip.png)](https://www.gittip.com/sjwhitworth/)
 
@@ -39,7 +40,7 @@ func main() {
 	// Load in a dataset, with headers. Header attributes will be stored.
 	// Think of instances as a Data Frame structure in R or Pandas.
 	// You can also create instances from scratch.
-	rawData, err := base.ParseCSVToInstances("datasets/iris.csv", false)
+	rawData, err := base.ParseCSVToInstances("datasets/iris.csv", true)
 	if err != nil {
 		panic(err)
 	}

base/csv.go

@@ -187,6 +187,11 @@ func ParseCSVBuildInstancesFromReader(r io.ReadSeeker, attrs []Attribute, hasHea
 			}
 		}
 		for i, v := range record {
+			// support missing values
+			if v == "" {
+				continue
+			}
+
 			u.Set(specs[i], rowCounter, specs[i].attr.GetSysValFromString(strings.TrimSpace(v)))
 		}
 		rowCounter++

base/dataframe_go.go

@@ -0,0 +1,66 @@
+package base
+
+import (
+	"fmt"
+	"reflect"
+	"strconv"
+
+	"github.com/rocketlaunchr/dataframe-go"
+)
+
+// ConvertDataFrameToInstances converts a DataFrame-go dataframe object to Golearn Fixed Data Grid. Allows for compabitibility between dataframe and golearn's ML models.
+// df is the dataframe Object. classAttrIndex is the index of the class Attribute in the data.i
+func ConvertDataFrameToInstances(df *dataframe.DataFrame, classAttrIndex int) FixedDataGrid {
+
+	// Creating Attributes based on Dataframe
+	names := df.Names()
+	attrs := make([]Attribute, len(names))
+
+	newInst := NewDenseInstances()
+
+	for i := range names {
+		col := df.Series[i]
+		if reflect.TypeOf(col.Value(0)).Kind() == reflect.String {
+			attrs[i] = new(CategoricalAttribute)
+			attrs[i].SetName(names[i])
+		} else {
+			attrs[i] = NewFloatAttribute(names[i])
+		}
+	}
+
+	// Add the attributes
+	newSpecs := make([]AttributeSpec, len(attrs))
+	for i, a := range attrs {
+
+		newSpecs[i] = newInst.AddAttribute(a)
+	}
+	// Adding the class attribute
+	newInst.AddClassAttribute(attrs[classAttrIndex])
+
+	// Allocate space
+	nRows := df.NRows()
+	newInst.Extend(df.NRows())
+
+	// Write the data based on DataType
+	for i := 0; i < nRows; i++ {
+		for j := range names {
+			col := df.Series[j]
+
+			var val string
+			switch v := col.Value(i).(type) {
+			case string:
+				val = v
+			case int64:
+				val = strconv.FormatInt(v, 10)
+			case float64:
+				val = fmt.Sprintf("%f", v)
+			case float32:
+				val = fmt.Sprintf("%f", v)
+			}
+
+			newInst.Set(newSpecs[j], i, newSpecs[j].GetAttribute().GetSysValFromString(val))
+		}
+	}
+
+	return newInst
+}

base/error.go

@@ -80,6 +80,6 @@ func WrapError(err error) error {
 }
 
 func FormatError(err error, format string, args ...interface{}) error {
-	description := fmt.Sprintf(format, args)
+	description := fmt.Sprintf(format, args...)
 	return DescribeError(description, err)
 }

base/error_test.go

@@ -0,0 +1,22 @@
+package base
+
+import (
+	. "github.com/smartystreets/goconvey/convey"
+	"testing"
+)
+
+func TestId3(t *testing.T) {
+	Convey("Doing a error test", t, func() {
+		var _gerr GoLearnError
+		gerr := &_gerr
+		gerr.attachFormattedStack()
+		s := gerr.Error()
+		So(s, ShouldNotBeNil)
+		err := DescribeError("test", nil)
+		So(err, ShouldNotBeNil)
+		err = WrapError(nil)
+		So(err, ShouldNotBeNil)
+		s = wrapLinesWithTabPrefix("123\ntest\n")
+		So(s, ShouldEqual, "\t123\n\ttest\n\t")
+	})
+}

base/mat_test.go

@@ -9,13 +9,13 @@ import (
 func TestInlineMat64Creation(t *testing.T) {
 
 	Convey("Given a literal array...", t, func() {
-		mat := mat.NewDense(4, 3, []float64{
+		X := mat.NewDense(4, 3, []float64{
 			1, 0, 1,
 			0, 1, 1,
 			0, 0, 0,
 			1, 1, 0,
 		})
-		inst := InstancesFromMat64(4, 3, mat)
+		inst := InstancesFromMat64(4, 3, X)
 		attrs := inst.AllAttributes()
 		Convey("Attributes should be well-defined...", func() {
 			So(len(attrs), ShouldEqual, 3)
@@ -34,6 +34,45 @@ func TestInlineMat64Creation(t *testing.T) {
 			So(val, ShouldAlmostEqual, 1.0)
 		})
 
+		Convey("Getting size should work...", func() {
+			attrLen, rows := inst.Size()
+			So(attrLen, ShouldEqual, 3)
+			So(rows, ShouldEqual, 4)
+		})
+
+		Convey("Getting row string should work...", func() {
+			So(inst.RowString(0), ShouldEqual, "0")
+		})
+
+		Convey("Getting attribute not in it should error...", func() {
+			Y := mat.NewDense(1, 4, []float64{1, 2, 3, 4})
+			ins := InstancesFromMat64(1, 4, Y)
+			attr := ins.AllAttributes()
+			_, err := inst.GetAttribute(attr[3])
+			So(err.Error(), ShouldEqual, "Couldn't find a matching attribute")
+		})
+
+		Convey("Generate human-readable summary...", func() {
+			output := inst.String()
+			So(output, ShouldStartWith, "Instances with")
+			So(output, ShouldContainSubstring, "Attributes:")
+			So(output, ShouldContainSubstring, "Data:")
+		})
+
 	})
 
 }
+
+func TestStringWithExceedMaxRow(t *testing.T) {
+	Convey("Given a long literal array...", t, func() {
+		v := make([]float64, 35, 35)
+		X := mat.NewDense(35, 1, v)
+		inst := InstancesFromMat64(35, 1, X)
+		output := inst.String()
+		So(output, ShouldStartWith, "Instances with")
+		So(output, ShouldContainSubstring, "Attributes:")
+		So(output, ShouldContainSubstring, "Data:")
+		So(output, ShouldContainSubstring, "undisplayed")
+
+	})
+}

base/serialize.go

@@ -49,17 +49,17 @@ func (f *FunctionalTarReader) GetNamedFile(name string) ([]byte, error) {
 			if err != nil {
 				return nil, WrapError(err)
 			}
+
 			if int64(len(ret)) != hdr.Size {
 				if int64(len(ret)) < hdr.Size {
 					log.Printf("Size mismatch, got %d byte(s) for %s, expected %d (err was %s)", len(ret), hdr.Name, hdr.Size, err)
 				} else {
 					return nil, WrapError(fmt.Errorf("Size mismatch, expected %d byte(s) for %s, got %d", len(ret), hdr.Name, hdr.Size))
 				}
 			}
-			if err != nil {
-				return nil, err
-			}
+
 			returnCandidate = ret
+			break
 		}
 	}
 	if returnCandidate == nil {

clustering/cluster_extra_test.go

@@ -0,0 +1,112 @@
+package clustering
+
+import (
+	. "github.com/smartystreets/goconvey/convey"
+	"testing"
+)
+
+func Test(t *testing.T) {
+	Convey("Only m[0]", t, func() {
+		m1 := ClusterMap(make(map[int][]int))
+		m1[0] = []int{1, 2}
+
+		m2 := ClusterMap(make(map[int][]int))
+		m2[0] = []int{1, 2}
+
+		ret, err := m1.Equals(m2)
+		So(err, ShouldBeNil)
+		So(ret, ShouldBeTrue)
+
+	})
+
+	Convey("Nothing in m", t, func() {
+		m1 := ClusterMap(make(map[int][]int))
+
+		m2 := ClusterMap(make(map[int][]int))
+
+		ret, err := m1.Equals(m2)
+		So(err, ShouldBeNil)
+		So(ret, ShouldBeTrue)
+
+	})
+
+	Convey("Many elements in m", t, func() {
+		m1 := ClusterMap(make(map[int][]int))
+		m1[0] = []int{1, 2, 3, 4, 5}
+		m1[1] = []int{11, 12, 13, 14, 15}
+
+		m2 := ClusterMap(make(map[int][]int))
+		m2[0] = []int{1, 2, 3, 4, 5}
+		m2[1] = []int{11, 12, 13, 14, 15}
+
+		ret, err := m1.Equals(m2)
+		So(err, ShouldBeNil)
+		So(ret, ShouldBeTrue)
+
+	})
+
+	Convey("m[0] not the same", t, func() {
+		m1 := ClusterMap(make(map[int][]int))
+		m1[1] = []int{1, 2, 3}
+		m1[0] = []int{4, 5}
+
+		m2 := ClusterMap(make(map[int][]int))
+		m2[1] = []int{1, 2, 3}
+		m2[0] = []int{6, 5}
+
+		_, err := m1.Equals(m2)
+		So(err, ShouldNotBeNil)
+	})
+
+	Convey("m[0] size diff", t, func() {
+		m1 := ClusterMap(make(map[int][]int))
+		m1[1] = []int{1, 2, 3}
+		m1[0] = []int{4, 5}
+
+		m2 := ClusterMap(make(map[int][]int))
+		m2[1] = []int{1, 2, 3}
+
+		_, err := m1.Equals(m2)
+		So(err, ShouldNotBeNil)
+	})
+
+	Convey("m[1] size diff", t, func() {
+		m1 := ClusterMap(make(map[int][]int))
+		m1[1] = []int{1, 3}
+		m1[0] = []int{4, 5}
+
+		m2 := ClusterMap(make(map[int][]int))
+		m2[1] = []int{1, 2, 3}
+		m1[0] = []int{4, 5}
+
+		_, err := m1.Equals(m2)
+		So(err, ShouldNotBeNil)
+	})
+
+	Convey("m[1] duplicate", t, func() {
+		m1 := ClusterMap(make(map[int][]int))
+		m1[1] = []int{1, 1}
+		m1[0] = []int{4, 5}
+
+		m2 := ClusterMap(make(map[int][]int))
+		m1[1] = []int{1, 1}
+		m1[0] = []int{4, 5}
+
+		_, err := m1.Equals(m2)
+		So(err, ShouldNotBeNil)
+	})
+
+	Convey("m[0] duplicate", t, func() {
+		m1 := ClusterMap(make(map[int][]int))
+		m1[1] = []int{1, 2}
+		m1[0] = []int{4, 4}
+
+		m2 := ClusterMap(make(map[int][]int))
+		m1[1] = []int{1, 2}
+		m1[0] = []int{4, 4}
+
+		_, err := m1.Equals(m2)
+		So(err, ShouldNotBeNil)
+	})
+
+}

clustering/dbscan.go

@@ -1,9 +1,9 @@
 package clustering
 
 import (
-	"gonum.org/v1/gonum/mat"
 	"github.com/sjwhitworth/golearn/base"
 	"github.com/sjwhitworth/golearn/metrics/pairwise"
+	"gonum.org/v1/gonum/mat"
 	"math/big"
 )
 

clustering/dbscan_test.go

@@ -2,10 +2,10 @@ package clustering
 
 import (
 	"bufio"
-	"gonum.org/v1/gonum/mat"
 	"github.com/sjwhitworth/golearn/base"
 	"github.com/sjwhitworth/golearn/metrics/pairwise"
 	. "github.com/smartystreets/goconvey/convey"
+	"gonum.org/v1/gonum/mat"
 	"math"
 	"math/big"
 	"os"