Updated v.2.2.1

Renamed deltaValue to deltaWeight.

KRMLP.podspec

@@ -1,6 +1,6 @@
 Pod::Spec.new do |s|
   s.name         = "KRMLP"
-  s.version      = "2.2.0"
+  s.version      = "2.2.1"
   s.summary      = "Deep Learning for multi-layer perceptrons neural network (MLP)."
   s.description  = <<-DESC
                    Machine Learning (マシンラーニング) in this project, it implemented multi-layer perceptrons neural network (ニューラルネットワーク) and Back Propagation Neural Network (BPN). It designed unlimited hidden layers to do the training tasks. This network can be used in products recommendation (おすすめの商品), user behavior analysis (ユーザーの行動分析), data mining (データマイニング) and data analysis (データ分析).
@@ -10,7 +10,7 @@ Pod::Spec.new do |s|
   s.author       = { "Kalvar Lin" => "[email protected]" }
   s.social_media_url = "https://twitter.com/ilovekalvar"
   s.source       = { :git => "https://github.com/Kalvar/ios-Multi-Perceptron-NeuralNetwork.git", :tag => s.version.to_s }
-  s.platform     = :ios, '7.0'
+  s.platform     = :ios, '9.0'
   s.requires_arc = true
   s.public_header_files = 'ML/**/*.h'
   s.source_files = 'ML/**/*.{h,m}'

LICENSE

@@ -1,4 +1,4 @@
-Copyright (c) 2013 - 2016 Kuo-Ming Lin (Kalvar Lin)
+Copyright (c) 2013 - 2017 Kuo-Ming Lin (Kalvar Lin)
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

ML/Optimizations/Inertia/KRMLPInertia.m

@@ -39,7 +39,7 @@ - (instancetype)init
 
 - (double)deltaWeightAtIndex:(NSInteger)weightIndex net:(KRMLPNet *)net mappedOutput:(double)mappedOutput learningRate:(double)learningRate
 {
-    double deltaValue = learningRate * net.deltaValue * mappedOutput;
+    double deltaWeight = learningRate * net.deltaValue * mappedOutput;
     if( net.updatedTimes > 0 )
     {
         double lastDeltaWeight = [[net.lastDeltaWeights objectAtIndex:weightIndex] doubleValue];
@@ -71,24 +71,24 @@ - (double)deltaWeightAtIndex:(NSInteger)weightIndex net:(KRMLPNet *)net mappedOu
                 {
                     dynamicLearningRate = 0.0f;
                 }
-                deltaValue += dynamicLearningRate * lastDeltaWeight;
+                deltaWeight += dynamicLearningRate * lastDeltaWeight;
             }
                 break;
             case KRMLPInertialRProp: // Todolist, the RProp is better than QuickProp, since QuickProp has overfitting problem.
             case KRMLPInertialFixedRate:
             default:
                 // delta w(ji) = L * delta value * y + fixed inertial rate * last delta w(ji)
-                deltaValue += _inertialRate * lastDeltaWeight;
+                deltaWeight += _inertialRate * lastDeltaWeight;
                 break;
         }
     }
     else
     {
         // 是初次更新權重的狀態
         // 直接照 Backpropagation 原始未套入優化公式的方式去做即可
-        // deltaValue = learningRate * net.deltaValue * mappedOutput;
+        // deltaWeight = learningRate * net.deltaValue * mappedOutput;
     }
-    return deltaValue;
+    return deltaWeight;
 }
 
 @end

README.md

@@ -6,8 +6,8 @@ Machine Learning (マシンラーニング) in this project, it implemented mult
 #### Podfile
 
 ```ruby
-platform :ios, '7.0'
-pod "KRMLP", "~> 2.2.0"
+platform :ios, '9.0'
+pod "KRMLP", "~> 2.2.1"
 ```
 
 ## How to use
@@ -275,15 +275,15 @@ QuickProp:
 
 ## Version
 
-V2.2.0
+V2.2.1
 
 ## License
 
 MIT.
 
 ## Todolist
 
-1. RProp.
-2. Mixes fixed inertia and QuickProp.
+1. RMSProp.
+2. Adam.
 3. EDBD.
-4. Protocol implementations.
+4. Nadam.