diff --git a/README.md b/README.md
index c55a4f2..a27f7ed 100644
--- a/README.md
+++ b/README.md
@@ -17,13 +17,15 @@ All the functions can be seen in [examples/main.go](https://github.com/montanafl
 
 ```go
 // start with some source data to use
-data := []float64{1, 2, 3, 4, 4, 5}
+data := []float64{1.0, 2.1, 3.2, 4.823, 4.1, 5.8}
 
-// you could also use different number types like this
-// data := stats.LoadRawData([]interface{}{1.1, "2", 3.0, 4, "5"})
+// you could also use different types like this
+// data := stats.LoadRawData([]int{1, 2, 3, 4, 5})
+// data := stats.LoadRawData([]interface{}{1.1, "2", 3})
+// etc...
 
 median, _ := stats.Median(data)
-fmt.Println(median) // 3.5
+fmt.Println(median) // 3.65
 
 roundedMedian, _ := stats.Round(median, 0)
 fmt.Println(roundedMedian) // 4
@@ -90,6 +92,8 @@ func PopulationVariance(input Float64Data) (pvar float64, err error) {}
 func Round(input float64, places int) (rounded float64, err error) {}
 func Sample(input Float64Data, takenum int, replacement bool) ([]float64, error) {}
 func SampleVariance(input Float64Data) (svar float64, err error) {}
+func Sigmoid(input Float64Data) ([]float64, error) {}
+func SoftMax(input Float64Data) ([]float64, error) {}
 func StandardDeviation(input Float64Data) (sdev float64, err error) {}
 func StandardDeviationPopulation(input Float64Data) (sdev float64, err error) {}
 func StandardDeviationSample(input Float64Data) (sdev float64, err error) {}
diff --git a/examples/main.go b/examples/main.go
index d4cd56f..e6044ef 100644
--- a/examples/main.go
+++ b/examples/main.go
@@ -8,7 +8,8 @@ import (
 
 func main() {
 
-	d := stats.LoadRawData([]interface{}{1.1, "2", 3.0, 4, "5"})
+	// d := stats.LoadRawData([]interface{}{1.1, "2", 3.0, 4, "5"})
+	d := stats.LoadRawData([]int{1, 2, 3, 4, 5})
 
 	a, _ := stats.Min(d)
 	fmt.Println(a)
@@ -156,4 +157,12 @@ func main() {
 	ac, _ := stats.AutoCorrelation([]float64{1, 2, 3, 4, 5}, 1)
 	fmt.Println(ac)
 	// Output: 0.4
+
+	sig, _ := stats.Sigmoid([]float64{3.0, 1.0, 0.2})
+	fmt.Println(s)
+	// Output: [0.9525741268224334 0.7310585786300049 0.549833997312478]
+
+	sm, _ := stats.SoftMax([]float64{3.0, 1.0, 0.2})
+	fmt.Println(sm)
+	// Output: [0.8360188027814407 0.11314284146556013 0.05083835575299916]
 }
diff --git a/sigmoid.go b/sigmoid.go
new file mode 100644
index 0000000..5f2559d
--- /dev/null
+++ b/sigmoid.go
@@ -0,0 +1,18 @@
+package stats
+
+import "math"
+
+// Sigmoid returns the input values in the range of -1 to 1
+// along the sigmoid or s-shaped curve, commonly used in
+// machine learning while training neural networks as an
+// activation function.
+func Sigmoid(input Float64Data) ([]float64, error) {
+	if input.Len() == 0 {
+		return Float64Data{}, EmptyInput
+	}
+	s := make([]float64, len(input))
+	for i, v := range input {
+		s[i] = 1 / (1 + math.Exp(-v))
+	}
+	return s, nil
+}
diff --git a/sigmoid_test.go b/sigmoid_test.go
new file mode 100644
index 0000000..bef9dbd
--- /dev/null
+++ b/sigmoid_test.go
@@ -0,0 +1,41 @@
+package stats
+
+import (
+	"fmt"
+	"testing"
+)
+
+func ExampleSigmoid() {
+	s, _ := Sigmoid([]float64{3.0, 1.0, 0.2})
+	fmt.Println(s)
+	// Output: [0.9525741268224334 0.7310585786300049 0.549833997312478]
+}
+
+func TestSigmoidEmptyInput(t *testing.T) {
+	_, err := Sigmoid([]float64{})
+	if err != EmptyInputErr {
+		t.Errorf("Should have returned empty input error")
+	}
+}
+
+func TestSigmoid(t *testing.T) {
+	sm, err := Sigmoid([]float64{-0.54761371, 17.04850603, 4.86054302})
+	if err != nil {
+		t.Error(err)
+	}
+
+	a := 0.3664182235138545
+	if sm[0] != a {
+		t.Errorf("%v != %v", sm[0], a)
+	}
+
+	a = 0.9999999605608187
+	if sm[1] != a {
+		t.Errorf("%v != %v", sm[1], a)
+	}
+
+	a = 0.9923132671908277
+	if sm[2] != a {
+		t.Errorf("%v != %v", sm[2], a)
+	}
+}
diff --git a/softmax.go b/softmax.go
new file mode 100644
index 0000000..8507264
--- /dev/null
+++ b/softmax.go
@@ -0,0 +1,25 @@
+package stats
+
+import "math"
+
+// SoftMax returns the input values in the range of 0 to 1
+// with sum of all the probabilities being equal to one. It
+// is commonly used in machine learning neural networks.
+func SoftMax(input Float64Data) ([]float64, error) {
+	if input.Len() == 0 {
+		return Float64Data{}, EmptyInput
+	}
+
+	s := 0.0
+	c, _ := Max(input)
+	for _, e := range input {
+		s += math.Exp(e - c)
+	}
+
+	sm := make([]float64, len(input))
+	for i, v := range input {
+		sm[i] = math.Exp(v-c) / s
+	}
+
+	return sm, nil
+}
diff --git a/softmax_test.go b/softmax_test.go
new file mode 100644
index 0000000..9b08224
--- /dev/null
+++ b/softmax_test.go
@@ -0,0 +1,41 @@
+package stats
+
+import (
+	"fmt"
+	"testing"
+)
+
+func ExampleSoftMax() {
+	sm, _ := SoftMax([]float64{3.0, 1.0, 0.2})
+	fmt.Println(sm)
+	// Output: [0.8360188027814407 0.11314284146556013 0.05083835575299916]
+}
+
+func TestSoftMaxEmptyInput(t *testing.T) {
+	_, err := SoftMax([]float64{})
+	if err != EmptyInputErr {
+		t.Errorf("Should have returned empty input error")
+	}
+}
+
+func TestSoftMax(t *testing.T) {
+	sm, err := SoftMax([]float64{3.0, 1.0, 0.2})
+	if err != nil {
+		t.Error(err)
+	}
+
+	a := 0.8360188027814407
+	if sm[0] != a {
+		t.Errorf("%v != %v", sm[0], a)
+	}
+
+	a = 0.11314284146556013
+	if sm[1] != a {
+		t.Errorf("%v != %v", sm[1], a)
+	}
+
+	a = 0.05083835575299916
+	if sm[2] != a {
+		t.Errorf("%v != %v", sm[1], a)
+	}
+}