* fixed he and xavier initializations

* added normal and uniform xavier init

src/main/java/net/echo/brain4j/model/initialization/WeightInit.java

@@ -1,9 +1,6 @@
 package net.echo.brain4j.model.initialization;
 
-import net.echo.brain4j.model.initialization.impl.HeInit;
-import net.echo.brain4j.model.initialization.impl.LeCunInit;
-import net.echo.brain4j.model.initialization.impl.NormalInit;
-import net.echo.brain4j.model.initialization.impl.XavierInit;
+import net.echo.brain4j.model.initialization.impl.*;
 
 /**
  * Enum that defines the different types of weight initialization strategies used for neural networks.
@@ -24,10 +21,18 @@ public enum WeightInit {
     HE(new HeInit()),
 
     /**
-     * Xavier initialization (also known as Glorot initialization) is designed for layers with sigmoid or tanh activations.
-     * It initializes weights using a uniform distribution with a variance based on the number of input and output neurons.
+     * Uniform Xavier initialization is specifically designed for layers with sigmoid or tanh activations.
+     * It initializes weights using a uniform distribution,
+     * scaled by the square root of 6 divided by the sum of the number of input and output neurons.
      */
-    XAVIER(new XavierInit()),
+    UNIFORM_XAVIER(new UniformXavierInit()),
+
+    /**
+     * Normal Xavier initialization is specifically designed for layers with sigmoid or tanh activations.
+     * It initializes weights using a normal distribution,
+     * scaled by the square root of 2 divided by the number of input and output neurons.
+     */
+    NORMAL_XAVIER(new NormalXavierInit()),
 
     /**
      * LeCun initialization is specifically designed for layers with the sigmoid or tanh activation functions.

src/main/java/net/echo/brain4j/model/initialization/impl/HeInit.java

@@ -6,6 +6,6 @@ public class HeInit implements WeightInitializer {
 
     @Override
     public double getBound(int nIn, int nOut) {
-        return SQRT_OF_6 / Math.sqrt(nIn);
+        return Math.sqrt(2.0 / nIn);
     }
 }

src/main/java/net/echo/brain4j/model/initialization/impl/XavierInit.java → src/main/java/net/echo/brain4j/model/initialization/impl/NormalXavierInit.java

@@ -2,10 +2,10 @@
 
 import net.echo.brain4j.model.initialization.WeightInitializer;
 
-public class XavierInit implements WeightInitializer {
+public class NormalXavierInit implements WeightInitializer {
 
     @Override
     public double getBound(int nIn, int nOut) {
-        return SQRT_OF_6 / Math.sqrt(nIn + nOut);
+        return Math.sqrt(2.0 / (nIn + nOut));
     }
 }

src/main/java/net/echo/brain4j/model/initialization/impl/UniformXavierInit.java

@@ -0,0 +1,11 @@
+package net.echo.brain4j.model.initialization.impl;
+
+import net.echo.brain4j.model.initialization.WeightInitializer;
+
+public class UniformXavierInit implements WeightInitializer {
+
+    @Override
+    public double getBound(int nIn, int nOut) {
+        return Math.sqrt(6.0 / (nIn + nOut));
+    }
+}