// generateTrainingAttrs selects RandomFeatures number of base.Attributes from
// the provided base.Instances.
func (b *BaggedModel) generateTrainingAttrs(model int, from base.FixedDataGrid) []base.Attribute {
ret := make([]base.Attribute, 0)
attrs := base.NonClassAttributes(from)
if b.RandomFeatures == 0 {
ret = attrs
} else {
for {
if len(ret) >= b.RandomFeatures {
break
}
attrIndex := rand.Intn(len(attrs))
attr := attrs[attrIndex]
matched := false
for _, a := range ret {
if a.Equals(attr) {
matched = true
break
}
}
if !matched {
ret = append(ret, attr)
}
}
}
for _, a := range from.AllClassAttributes() {
ret = append(ret, a)
}
b.lock.Lock()
b.selectedAttributes[model] = ret
b.lock.Unlock()
return ret
}
func convertToBinary(src base.FixedDataGrid) base.FixedDataGrid {
// Convert to binary
b := filters.NewBinaryConvertFilter()
attrs := base.NonClassAttributes(src)
for _, a := range attrs {
b.AddAttribute(a)
}
b.Train()
ret := base.NewLazilyFilteredInstances(src, b)
return ret
}
func (KNN *KNNClassifier) canUseOptimisations(what base.FixedDataGrid) bool {
// Check that the two have exactly the same layout
if !base.CheckStrictlyCompatible(what, KNN.TrainingData) {
return false
}
// Check that the two are DenseInstances
whatd, ok1 := what.(*base.DenseInstances)
_, ok2 := KNN.TrainingData.(*base.DenseInstances)
if !ok1 || !ok2 {
return false
}
// Check that no Class Attributes are mixed in with the data
classAttrs := whatd.AllClassAttributes()
normalAttrs := base.NonClassAttributes(whatd)
// Retrieve all the AGs
ags := whatd.AllAttributeGroups()
classAttrGroups := make([]base.AttributeGroup, 0)
for agName := range ags {
ag := ags[agName]
attrs := ag.Attributes()
matched := false
for _, a := range attrs {
for _, c := range classAttrs {
if a.Equals(c) {
matched = true
}
}
}
if matched {
classAttrGroups = append(classAttrGroups, ag)
}
}
for _, cag := range classAttrGroups {
attrs := cag.Attributes()
common := base.AttributeIntersect(normalAttrs, attrs)
if len(common) != 0 {
return false
}
}
// Check that all of the Attributes are numeric
for _, a := range normalAttrs {
if _, ok := a.(*base.FloatAttribute); !ok {
return false
}
}
// If that's fine, return true
return true
}
func TestRandomForest(t *testing.T) {
Convey("Given a valid CSV file", t, func() {
inst, err := base.ParseCSVToInstances("../examples/datasets/iris_headers.csv", true)
So(err, ShouldBeNil)
Convey("When Chi-Merge filtering the data", func() {
filt := filters.NewChiMergeFilter(inst, 0.90)
for _, a := range base.NonClassFloatAttributes(inst) {
filt.AddAttribute(a)
}
filt.Train()
instf := base.NewLazilyFilteredInstances(inst, filt)
Convey("Splitting the data into test and training sets", func() {
trainData, testData := base.InstancesTrainTestSplit(instf, 0.60)
Convey("Fitting and predicting with a Random Forest", func() {
rf := NewRandomForest(10, 3)
err = rf.Fit(trainData)
So(err, ShouldBeNil)
predictions, err := rf.Predict(testData)
So(err, ShouldBeNil)
confusionMat, err := evaluation.GetConfusionMatrix(testData, predictions)
So(err, ShouldBeNil)
Convey("Predictions should be somewhat accurate", func() {
So(evaluation.GetAccuracy(confusionMat), ShouldBeGreaterThan, 0.35)
})
})
})
})
Convey("Fitting with a Random Forest with too many features compared to the data", func() {
rf := NewRandomForest(10, len(base.NonClassAttributes(inst))+1)
err = rf.Fit(inst)
Convey("Should return an error", func() {
So(err, ShouldNotBeNil)
})
})
})
}
// Fit builds the RandomForest on the specified instances
func (f *RandomForest) Fit(on base.FixedDataGrid) error {
numNonClassAttributes := len(base.NonClassAttributes(on))
if numNonClassAttributes < f.Features {
return errors.New(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++ {
tree := trees.NewID3DecisionTree(0.00)
f.Model.AddModel(tree)
}
f.Model.Fit(on)
return nil
}
func TestBinaryFilter(t *testing.T) {
Convey("Given a contrived dataset...", t, func() {
// Read the contrived dataset
inst, err := base.ParseCSVToInstances("./binary_test.csv", true)
So(err, ShouldEqual, nil)
// Add Attributes to the filter
bFilt := NewBinaryConvertFilter()
bAttrs := base.NonClassAttributes(inst)
for _, a := range bAttrs {
bFilt.AddAttribute(a)
}
bFilt.Train()
// Construct a LazilyFilteredInstances to handle it
instF := base.NewLazilyFilteredInstances(inst, bFilt)
Convey("All the non-class Attributes should be binary...", func() {
// Check that all the Attributes are the right type
for _, a := range base.NonClassAttributes(instF) {
_, ok := a.(*base.BinaryAttribute)
So(ok, ShouldEqual, true)
}
})
// Check that all the class Attributes made it
Convey("All the class Attributes should have survived...", func() {
origClassAttrs := inst.AllClassAttributes()
newClassAttrs := instF.AllClassAttributes()
intersectClassAttrs := base.AttributeIntersect(origClassAttrs, newClassAttrs)
So(len(intersectClassAttrs), ShouldEqual, len(origClassAttrs))
})
// Check that the Attributes have the right names
Convey("Attribute names should be correct...", func() {
origNames := []string{"floatAttr", "shouldBe1Binary",
"shouldBe3Binary_stoicism", "shouldBe3Binary_heroism",
"shouldBe3Binary_romanticism", "arbitraryClass"}
origMap := make(map[string]bool)
for _, a := range origNames {
origMap[a] = false
}
for _, a := range instF.AllAttributes() {
name := a.GetName()
_, ok := origMap[name]
if !ok {
t.Error(fmt.Sprintf("Weird: %s", name))
}
origMap[name] = true
}
for a := range origMap {
So(origMap[a], ShouldEqual, true)
}
})
// Check that the Attributes have been discretised correctly
Convey("Discretisation should have worked", func() {
// Build Attribute map
attrMap := make(map[string]base.Attribute)
for _, a := range instF.AllAttributes() {
attrMap[a.GetName()] = a
}
// For each attribute
for name := range attrMap {
attr := attrMap[name]
// Retrieve AttributeSpec
as, err := instF.GetAttribute(attr)
So(err, ShouldEqual, nil)
if name == "floatAttr" {
So(instF.Get(as, 0), ShouldResemble, []byte{1})
So(instF.Get(as, 1), ShouldResemble, []byte{1})
So(instF.Get(as, 2), ShouldResemble, []byte{0})
} else if name == "shouldBe1Binary" {
So(instF.Get(as, 0), ShouldResemble, []byte{0})
So(instF.Get(as, 1), ShouldResemble, []byte{1})
So(instF.Get(as, 2), ShouldResemble, []byte{1})
} else if name == "shouldBe3Binary_stoicism" {
So(instF.Get(as, 0), ShouldResemble, []byte{1})
So(instF.Get(as, 1), ShouldResemble, []byte{0})
So(instF.Get(as, 2), ShouldResemble, []byte{0})
} else if name == "shouldBe3Binary_heroism" {
So(instF.Get(as, 0), ShouldResemble, []byte{0})
So(instF.Get(as, 1), ShouldResemble, []byte{1})
So(instF.Get(as, 2), ShouldResemble, []byte{0})
} else if name == "shouldBe3Binary_romanticism" {
So(instF.Get(as, 0), ShouldResemble, []byte{0})
So(instF.Get(as, 1), ShouldResemble, []byte{0})
So(instF.Get(as, 2), ShouldResemble, []byte{1})
} else if name == "arbitraryClass" {
} else {
t.Error("Shouldn't have %s", name)
}
}
})
})
}
// Predict uses the underlying network to produce predictions for the
// class variables of X.
//
// Can only predict one CategoricalAttribute at a time, or up to n
// FloatAttributes. Set or unset ClassAttributes to work around this
// limitation.
func (m *MultiLayerNet) Predict(X base.FixedDataGrid) base.FixedDataGrid {
// Create the return vector
ret := base.GeneratePredictionVector(X)
// Make sure everything's a FloatAttribute
insts := m.convertToFloatInsts(X)
// Get the input/output Attributes
inputAttrs := base.NonClassAttributes(insts)
outputAttrs := ret.AllClassAttributes()
// Compute layers
layers := 2 + len(m.layers)
// Check that we're operating in a singular mode
floatMode := 0
categoricalMode := 0
for _, a := range outputAttrs {
if _, ok := a.(*base.CategoricalAttribute); ok {
categoricalMode++
} else if _, ok := a.(*base.FloatAttribute); ok {
floatMode++
} else {
panic("Unsupported output Attribute type!")
}
}
if floatMode > 0 && categoricalMode > 0 {
panic("Can't predict a mix of float and categorical Attributes")
} else if categoricalMode > 1 {
panic("Can't predict more than one categorical class Attribute")
}
// Create the activation vector
a := mat64.NewDense(m.network.size, 1, make([]float64, m.network.size))
// Resolve the input AttributeSpecs
inputAs := base.ResolveAttributes(insts, inputAttrs)
// Resolve the output Attributespecs
outputAs := base.ResolveAttributes(ret, outputAttrs)
// Map over each input row
insts.MapOverRows(inputAs, func(row [][]byte, rc int) (bool, error) {
// Clear the activation vector
for i := 0; i < m.network.size; i++ {
a.Set(i, 0, 0.0)
}
// Build the activation vector
for i, vb := range row {
if cIndex, ok := m.attrs[inputAs[i].GetAttribute()]; !ok {
panic("Can't resolve the Attribute!")
} else {
a.Set(cIndex, 0, base.UnpackBytesToFloat(vb))
}
}
// Robots, activate!
m.network.Activate(a, layers)
// Decide which class to set
if floatMode > 0 {
for _, as := range outputAs {
cIndex := m.attrs[as.GetAttribute()]
ret.Set(as, rc, base.PackFloatToBytes(a.At(cIndex, 0)))
}
} else {
maxIndex := 0
maxVal := 0.0
for i := m.classAttrOffset; i < m.classAttrOffset+m.classAttrCount; i++ {
val := a.At(i, 0)
if val > maxVal {
maxIndex = i
maxVal = val
}
}
maxIndex -= m.classAttrOffset
ret.Set(outputAs[0], rc, base.PackU64ToBytes(uint64(maxIndex)))
}
return true, nil
})
return ret
}
// Fit trains the neural network on the given fixed datagrid.
//
// Training stops when the mean-squared error acheived is less
// than the Convergence value, or when back-propagation has occured
// more times than the value set by MaxIterations.
func (m *MultiLayerNet) Fit(X base.FixedDataGrid) {
// Make sure everything's a FloatAttribute
insts := m.convertToFloatInsts(X)
// The size of the first layer is the number of things
// in the revised instances which aren't class Attributes
inputAttrsVec := base.NonClassAttributes(insts)
// The size of the output layer is the number of things
// in the revised instances which are class Attributes
classAttrsVec := insts.AllClassAttributes()
// The total number of layers is input layer + output layer
// plus number of layers specified
totalLayers := 2 + len(m.layers)
// The size is then augmented by the number of nodes
// in the centre
size := len(inputAttrsVec)
size += len(classAttrsVec)
hiddenSize := 0
for _, a := range m.layers {
size += a
hiddenSize += a
}
// Enumerate the Attributes
trainingAttrs := make(map[base.Attribute]int)
classAttrs := make(map[base.Attribute]int)
attrCounter := 0
for i, a := range inputAttrsVec {
attrCounter = i
m.attrs[a] = attrCounter
trainingAttrs[a] = attrCounter
}
m.classAttrOffset = attrCounter + 1
for _, a := range classAttrsVec {
attrCounter++
m.attrs[a] = attrCounter + hiddenSize
classAttrs[a] = attrCounter + hiddenSize
m.classAttrCount++
}
// Create the underlying Network
m.network = NewNetwork(size, len(inputAttrsVec), Sigmoid)
// Initialise inter-hidden layer weights and biases to small random values
layerOffset := len(inputAttrsVec)
for i := 0; i < len(m.layers)-1; i++ {
// Get the size of this layer
thisLayerSize := m.layers[i]
// Next layer size
nextLayerSize := m.layers[i+1]
// For every node in this layer
for j := 1; j <= thisLayerSize; j++ {
// Compute the offset
nodeOffset1 := layerOffset + j
// For every node in the next layer
for k := 1; k <= nextLayerSize; k++ {
// Compute offset
nodeOffset2 := layerOffset + thisLayerSize + k
// Set weight randomly
m.network.SetWeight(nodeOffset1, nodeOffset2, rand.NormFloat64()*0.1)
}
}
layerOffset += thisLayerSize
}
// Initialise biases with each hidden layer
layerOffset = len(inputAttrsVec)
for _, l := range m.layers {
for j := 1; j <= l; j++ {
nodeOffset := layerOffset + j
m.network.SetBias(nodeOffset, rand.NormFloat64()*0.1)
}
layerOffset += l
}
// Initialise biases for output layer
for i := 0; i < len(classAttrsVec); i++ {
nodeOffset := layerOffset + i
m.network.SetBias(nodeOffset, rand.NormFloat64()*0.1)
}
// Connect final hidden layer with the output layer
layerOffset = len(inputAttrsVec)
for i, l := range m.layers {
if i == len(m.layers)-1 {
for j := 1; j <= l; j++ {
nodeOffset1 := layerOffset + j
for k := 1; k <= len(classAttrsVec); k++ {
nodeOffset2 := layerOffset + l + k
m.network.SetWeight(nodeOffset1, nodeOffset2, rand.NormFloat64()*0.1)
}
}
}
layerOffset += l
}
// Connect input layer with first hidden layer (or output layer
for i := 1; i <= len(inputAttrsVec); i++ {
nextLayerLen := 0
if len(m.layers) > 0 {
nextLayerLen = m.layers[0]
} else {
nextLayerLen = len(classAttrsVec)
}
for j := 1; j <= nextLayerLen; j++ {
nodeOffset := len(inputAttrsVec) + j
v := rand.NormFloat64() * 0.1
m.network.SetWeight(i, nodeOffset, v)
}
}
// Create the training activation vector
trainVec := mat64.NewDense(size, 1, make([]float64, size))
// Create the error vector
errVec := mat64.NewDense(size, 1, make([]float64, size))
// Resolve training AttributeSpecs
trainAs := base.ResolveAllAttributes(insts)
// Feed-forward, compute error and update for each training example
// until convergence (what's that)
for iteration := 0; iteration < m.MaxIterations; iteration++ {
totalError := 0.0
maxRow := 0
insts.MapOverRows(trainAs, func(row [][]byte, i int) (bool, error) {
maxRow = i
// Clear vectors
for i := 0; i < size; i++ {
trainVec.Set(i, 0, 0.0)
errVec.Set(i, 0, 0.0)
}
// Build vectors
for i, vb := range row {
v := base.UnpackBytesToFloat(vb)
if attrIndex, ok := trainingAttrs[trainAs[i].GetAttribute()]; ok {
// Add to Activation vector
trainVec.Set(attrIndex, 0, v)
} else if attrIndex, ok := classAttrs[trainAs[i].GetAttribute()]; ok {
// Set to error vector
errVec.Set(attrIndex, 0, v)
} else {
panic("Should be able to find this Attribute!")
}
}
// Activate the network
m.network.Activate(trainVec, totalLayers-1)
// Compute the error
for a := range classAttrs {
cIndex := classAttrs[a]
errVec.Set(cIndex, 0, errVec.At(cIndex, 0)-trainVec.At(cIndex, 0))
}
// Update total error
totalError += math.Abs(errVec.Sum())
// Back-propagate the error
b := m.network.Error(trainVec, errVec, totalLayers)
// Update the weights
m.network.UpdateWeights(trainVec, b, m.LearningRate)
// Update the biases
m.network.UpdateBias(b, m.LearningRate)
return true, nil
})
totalError /= float64(maxRow)
// If we've converged, no need to carry on
if totalError < m.Convergence {
break
}
}
}
func TestFloatFilter(t *testing.T) {
Convey("Given a contrived dataset...", t, func() {
// Read the contrived dataset
inst, err := base.ParseCSVToInstances("./binary_test.csv", true)
So(err, ShouldEqual, nil)
// Add Attributes to the filter
bFilt := NewFloatConvertFilter()
bAttrs := base.NonClassAttributes(inst)
for _, a := range bAttrs {
bFilt.AddAttribute(a)
}
bFilt.Train()
// Construct a LazilyFilteredInstances to handle it
instF := base.NewLazilyFilteredInstances(inst, bFilt)
Convey("All the non-class Attributes should be floats...", func() {
// Check that all the Attributes are the right type
for _, a := range base.NonClassAttributes(instF) {
_, ok := a.(*base.FloatAttribute)
So(ok, ShouldEqual, true)
}
})
// Check that all the class Attributes made it
Convey("All the class Attributes should have survived...", func() {
origClassAttrs := inst.AllClassAttributes()
newClassAttrs := instF.AllClassAttributes()
intersectClassAttrs := base.AttributeIntersect(origClassAttrs, newClassAttrs)
So(len(intersectClassAttrs), ShouldEqual, len(origClassAttrs))
})
// Check that the Attributes have the right names
Convey("Attribute names should be correct...", func() {
origNames := []string{"floatAttr", "shouldBe1Binary",
"shouldBe3Binary_stoicism", "shouldBe3Binary_heroism",
"shouldBe3Binary_romanticism", "arbitraryClass"}
origMap := make(map[string]bool)
for _, a := range origNames {
origMap[a] = false
}
for _, a := range instF.AllAttributes() {
name := a.GetName()
_, ok := origMap[name]
So(ok, ShouldBeTrue)
origMap[name] = true
}
for a := range origMap {
So(origMap[a], ShouldEqual, true)
}
})
Convey("All Attributes should be the correct type...", func() {
for _, a := range instF.AllAttributes() {
if a.GetName() == "arbitraryClass" {
_, ok := a.(*base.CategoricalAttribute)
So(ok, ShouldEqual, true)
} else {
_, ok := a.(*base.FloatAttribute)
So(ok, ShouldEqual, true)
}
}
})
// Check that the Attributes have been discretised correctly
Convey("FloatConversion should have worked", func() {
// Build Attribute map
attrMap := make(map[string]base.Attribute)
for _, a := range instF.AllAttributes() {
attrMap[a.GetName()] = a
}
// For each attribute
for name := range attrMap {
So(name, ShouldBeIn, []string{
"floatAttr",
"shouldBe1Binary",
"shouldBe3Binary_stoicism",
"shouldBe3Binary_heroism",
"shouldBe3Binary_romanticism",
"arbitraryClass",
})
attr := attrMap[name]
as, err := instF.GetAttribute(attr)
So(err, ShouldEqual, nil)
if name == "floatAttr" {
So(instF.Get(as, 0), ShouldResemble, base.PackFloatToBytes(1.0))
So(instF.Get(as, 1), ShouldResemble, base.PackFloatToBytes(1.0))
So(instF.Get(as, 2), ShouldResemble, base.PackFloatToBytes(0.0))
} else if name == "shouldBe1Binary" {
So(instF.Get(as, 0), ShouldResemble, base.PackFloatToBytes(0.0))
So(instF.Get(as, 1), ShouldResemble, base.PackFloatToBytes(1.0))
So(instF.Get(as, 2), ShouldResemble, base.PackFloatToBytes(1.0))
} else if name == "shouldBe3Binary_stoicism" {
So(instF.Get(as, 0), ShouldResemble, base.PackFloatToBytes(1.0))
So(instF.Get(as, 1), ShouldResemble, base.PackFloatToBytes(0.0))
So(instF.Get(as, 2), ShouldResemble, base.PackFloatToBytes(0.0))
} else if name == "shouldBe3Binary_heroism" {
So(instF.Get(as, 0), ShouldResemble, base.PackFloatToBytes(0.0))
So(instF.Get(as, 1), ShouldResemble, base.PackFloatToBytes(1.0))
So(instF.Get(as, 2), ShouldResemble, base.PackFloatToBytes(0.0))
} else if name == "shouldBe3Binary_romanticism" {
So(instF.Get(as, 0), ShouldResemble, base.PackFloatToBytes(0.0))
So(instF.Get(as, 1), ShouldResemble, base.PackFloatToBytes(0.0))
So(instF.Get(as, 2), ShouldResemble, base.PackFloatToBytes(1.0))
} else if name == "arbitraryClass" {
}
}
})
})
}
func (lr *LinearRegression) Fit(inst base.FixedDataGrid) error {
// Retrieve row size
_, rows := inst.Size()
// Validate class Attribute count
classAttrs := inst.AllClassAttributes()
if len(classAttrs) != 1 {
return fmt.Errorf("Only 1 class variable is permitted")
}
classAttrSpecs := base.ResolveAttributes(inst, classAttrs)
// Retrieve relevant Attributes
allAttrs := base.NonClassAttributes(inst)
attrs := make([]base.Attribute, 0)
for _, a := range allAttrs {
if _, ok := a.(*base.FloatAttribute); ok {
attrs = append(attrs, a)
}
}
cols := len(attrs) + 1
if rows < cols {
return NotEnoughDataError
}
// Retrieve relevant Attribute specifications
attrSpecs := base.ResolveAttributes(inst, attrs)
// Split into two matrices, observed results (dependent variable y)
// and the explanatory variables (X) - see http://en.wikipedia.org/wiki/Linear_regression
observed := mat64.NewDense(rows, 1, nil)
explVariables := mat64.NewDense(rows, cols, nil)
// Build the observed matrix
inst.MapOverRows(classAttrSpecs, func(row [][]byte, i int) (bool, error) {
val := base.UnpackBytesToFloat(row[0])
observed.Set(i, 0, val)
return true, nil
})
// Build the explainatory variables
inst.MapOverRows(attrSpecs, func(row [][]byte, i int) (bool, error) {
// Set intercepts to 1.0
explVariables.Set(i, 0, 1.0)
for j, r := range row {
explVariables.Set(i, j+1, base.UnpackBytesToFloat(r))
}
return true, nil
})
n := cols
qr := new(mat64.QR)
qr.Factorize(explVariables)
var q, reg mat64.Dense
q.QFromQR(qr)
reg.RFromQR(qr)
var transposed, qty mat64.Dense
transposed.Clone(q.T())
qty.Mul(&transposed, observed)
regressionCoefficients := make([]float64, n)
for i := n - 1; i >= 0; i-- {
regressionCoefficients[i] = qty.At(i, 0)
for j := i + 1; j < n; j++ {
regressionCoefficients[i] -= regressionCoefficients[j] * reg.At(i, j)
}
regressionCoefficients[i] /= reg.At(i, i)
}
lr.disturbance = regressionCoefficients[0]
lr.regressionCoefficients = regressionCoefficients[1:]
lr.fitted = true
lr.attrs = attrs
lr.cls = classAttrs[0]
return nil
}