OrgLion-ML/golearn
1
0
Fork 0
mirror of https://github.com/sjwhitworth/golearn.git synced 2025-04-25 13:48:49 +08:00
Code Issues Releases Wiki Activity

Format code

Browse Source
This commit is contained in:
ss8651twtw 2018-06-16 22:14:18 +08:00
parent 9bfe206c9c
commit 1e1b5f11fb
5 changed files with 82 additions and 85 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
base/error_test.go
View File

@ -1,12 +1,12 @@
package base
import (
. "github.com/smartystreets/goconvey/convey"
"testing"
)
func TestId3(t *testing.T) {
Convey("Doing a error test", t, func(){
Convey("Doing a error test", t, func() {
var _gerr GoLearnError
gerr := &_gerr
gerr.attachFormattedStack()

144
knn/knn_cov_test.go
View File

@ -9,86 +9,86 @@ import (
func TestKnnClassifierCov(t *testing.T) {
Convey("Test predict", t, func() {
Convey("distance function", func(){
trainingData, err := base.ParseCSVToInstances("knn_train_1.csv", false)
So(err, ShouldBeNil)
Convey("distance function", func() {
trainingData, err := base.ParseCSVToInstances("knn_train_1.csv", false)
So(err, ShouldBeNil)
testingData, err := base.ParseCSVToInstances("knn_test_1.csv", false)
So(err, ShouldBeNil)
testingData, err := base.ParseCSVToInstances("knn_test_1.csv", false)
So(err, ShouldBeNil)
Convey("use euclidean", func(){
cls := NewKnnClassifier("euclidean", "kdtree", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(err, ShouldBeNil)
So(predictions, ShouldNotEqual, nil)
result := base.GetClass(predictions, 0)
So(result, ShouldEqual, "blue")
})
Convey("use euclidean", func() {
cls := NewKnnClassifier("euclidean", "kdtree", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(err, ShouldBeNil)
So(predictions, ShouldNotEqual, nil)
result := base.GetClass(predictions, 0)
So(result, ShouldEqual, "blue")
})
Convey("use manhattan", func(){
cls := NewKnnClassifier("manhattan", "kdtree", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(err, ShouldBeNil)
So(predictions, ShouldNotEqual, nil)
result := base.GetClass(predictions, 0)
So(result, ShouldEqual, "blue")
})
Convey("use manhattan", func() {
cls := NewKnnClassifier("manhattan", "kdtree", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(err, ShouldBeNil)
So(predictions, ShouldNotEqual, nil)
result := base.GetClass(predictions, 0)
So(result, ShouldEqual, "blue")
})
Convey("use cosine", func(){
cls := NewKnnClassifier("cosine", "kdtree", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(err, ShouldBeNil)
So(predictions, ShouldNotEqual, nil)
result := base.GetClass(predictions, 0)
So(result, ShouldEqual, "blue")
})
Convey("use cosine", func() {
cls := NewKnnClassifier("cosine", "kdtree", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(err, ShouldBeNil)
So(predictions, ShouldNotEqual, nil)
result := base.GetClass(predictions, 0)
So(result, ShouldEqual, "blue")
})
Convey("use undefined distance function", func(){
cls := NewKnnClassifier("abcd", "kdtree", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(predictions, ShouldBeNil)
So(err.Error(), ShouldEqual, "unsupported distance function")
})
})
Convey("use undefined distance function", func() {
cls := NewKnnClassifier("abcd", "kdtree", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(predictions, ShouldBeNil)
So(err.Error(), ShouldEqual, "unsupported distance function")
})
})
Convey("searching algorithm", func(){
trainingData, err := base.ParseCSVToInstances("knn_train_1.csv", false)
So(err, ShouldBeNil)
Convey("searching algorithm", func() {
trainingData, err := base.ParseCSVToInstances("knn_train_1.csv", false)
So(err, ShouldBeNil)
testingData, err := base.ParseCSVToInstances("knn_test_1.csv", false)
So(err, ShouldBeNil)
testingData, err := base.ParseCSVToInstances("knn_test_1.csv", false)
So(err, ShouldBeNil)
Convey("use undefined searching algorithm", func(){
cls := NewKnnClassifier("cosine", "abcd", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(predictions, ShouldBeNil)
So(err.Error(), ShouldEqual, "unsupported searching algorithm")
})
})
Convey("use undefined searching algorithm", func() {
cls := NewKnnClassifier("cosine", "abcd", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(predictions, ShouldBeNil)
So(err.Error(), ShouldEqual, "unsupported searching algorithm")
})
})
Convey("check features", func(){
Convey("use different dataset", func(){
trainingData, err := base.ParseCSVToInstances("knn_train_1.csv", false)
So(err, ShouldBeNil)
testingData, err := base.ParseCSVToInstances("knn_test_2.csv", false)
So(err, ShouldBeNil)
cls := NewKnnClassifier("cosine", "linear", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(predictions, ShouldBeNil)
So(err.Error(), ShouldEqual, "attributes not compatible")
})
})
Convey("check features", func() {
Convey("use different dataset", func() {
trainingData, err := base.ParseCSVToInstances("knn_train_1.csv", false)
So(err, ShouldBeNil)
testingData, err := base.ParseCSVToInstances("knn_test_2.csv", false)
So(err, ShouldBeNil)
cls := NewKnnClassifier("cosine", "linear", 2)
cls.AllowOptimisations = false
cls.Fit(trainingData)
predictions, err := cls.Predict(testingData)
So(predictions, ShouldBeNil)
So(err.Error(), ShouldEqual, "attributes not compatible")
})
})
})
}

9
linear_models/linearsvc_test.go
View File

@ -9,12 +9,11 @@ import (
//"fmt"
)
func TestLinearSVC(t *testing.T) {
Convey("Doing a LinearSVC test", t, func(){
Convey("Doing a LinearSVC test", t, func() {
var SVC *LinearSVC
var err error
Convey("Test NewLinearSVC", func(){
Convey("Test NewLinearSVC", func() {
_, err = NewLinearSVC("l1", "l1", false, 1.0, -1e6)
So(err, ShouldNotBeNil)
_, err = NewLinearSVC("l0", "l1", false, 1.0, -1e6)
@ -35,7 +34,7 @@ func TestLinearSVC(t *testing.T) {
_, err = NewLinearSVC("l2", "l1", true, 1.0, -1e6)
So(err, ShouldNotBeNil)
So(func(){ SVC.GetMetadata() } , ShouldNotPanic)
So(func() { SVC.GetMetadata() }, ShouldNotPanic)
params := &LinearSVCParams{0, []float64{0.0}, 1.0, -1e6, false, false}
params = params.Copy()
@ -60,7 +59,6 @@ func TestLinearSVC(t *testing.T) {
err = SVC.Load("tmp")
So(err, ShouldBeNil)
inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
inst.RemoveClassAttribute(inst.AllAttributes()[4])
inst.AddClassAttribute(inst.AllAttributes()[1])
@ -76,7 +74,6 @@ func TestLinearSVC(t *testing.T) {
})
//err = SVC.Save("tmp")
//var problem *Problem
//var param *Parameter
})

2
linear_models/logistic_test.go
View File

@ -7,7 +7,7 @@ import (
)
func TestLogistic(t *testing.T) {
Convey("Doing a logistic test", t, func(){
Convey("Doing a logistic test", t, func() {
X, err := base.ParseCSVToInstances("train.csv", false)
So(err, ShouldEqual, nil)
Y, err := base.ParseCSVToInstances("test.csv", false)

6
trees/id3_test.go
View File

@ -1,14 +1,14 @@
package trees
import (
"fmt"
"github.com/sjwhitworth/golearn/base"
. "github.com/smartystreets/goconvey/convey"
"testing"
"fmt"
)
func TestId3(t *testing.T) {
Convey("Doing a id3 test", t, func(){
Convey("Doing a id3 test", t, func() {
var rule DecisionTreeRule
s := rule.String()
So(s, ShouldNotBeNil)