KRANN 2.1.4

您要在您的 Podfile 中添加如下类似的内容：pod 'KRANN', '~> 2.1'

target 'MyApp' do
  pod 'KRANN', '~> 2.1'
end

然后在终端内部或从 CocoaPods.app 运行 pod install

或者要测试运行，运行以下命令

pod try KRANN

KRANN 2.1.4

作者
Kalvar Lin
ilovekalvar

ios-Multi-Perceptron-NeuralNetwork

本项目中实现了机器学习（マシンラーニング），基于反向传播神经网络（BPN）实现了多层感知器神经网络（ニューラルネットワーク），并设计了无限隐藏层用于训练任务，并预留了灵活的空间以供未来将模糊理论结合到网络中。该网络可用于产品推荐（おすすめの商品）、用户行为分析（ユーザーの行動分析）、数据挖掘（データマイニング）和数据分析（データ分析）。

本项目设计用于移动设备执行基本数据分析。如果您需要了解如何使用此网络，请通过电子邮件向我咨询。

Podfile

platform :ios, '7.0'
pod "KRANN", "~> 2.1.4"

如何使用

#import "KRANN.h"

@interface ViewController ()<KRANNDelegate>

@property (nonatomic, strong) KRANN *_krANN;

@end

@implementation ViewController

@synthesize _krANN;

//Setups any detail, and 2 outputs, you could set more outputs.
-(void)useSample1
{
    _krANN.activationFunction = KRANNActivationFunctionSigmoid;
    //各輸入向量陣列值 & 每一筆輸入向量的期望值( 輸出期望 )，因使用 S 形轉換函數，故 Input 值域須為 [0, 1]，輸出目標為 [0, 1]
    //Add the patterns, the weights connect with hidden layer, the output targets
    [_krANN addPatterns:@[@1, @0.1, @0.5, @0.2] outputGoals:@[@0.7f, @0.8f]];  //Pattern 1, net 1, 2, 3, 4, the output layer is 2 nets
    [_krANN addPatterns:@[@0, @1, @0.3, @0.9] outputGoals:@[@0.1f, @0.1f]];    //Pattern 2, same as pattern 1
    [_krANN addPatterns:@[@1, @0.3, @0.1, @0.4] outputGoals:@[@0.95f, @0.9f]]; //Pattern 3, same as pattern 1

    //輸入層各向量值到隱藏層神經元的權重 ( 連結同一個 Net 的就一組一組分開，有幾個 Hidden Net 就會有幾組 )
    //Add pattern-weights in Input layer to first Hidden Layer
    [_krANN addPatternWeights:@[@0.2, @-0.3]]; //W15, W16
    [_krANN addPatternWeights:@[@0.4, @0.1]];  //W25, W26
    [_krANN addPatternWeights:@[@-0.5, @0.2]]; //W35, W36
    [_krANN addPatternWeights:@[@-0.1, @0.3]]; //W45, W46

    //隱藏層神經元的偏權值 & 隱藏層神經元到輸出層神經元的權重值
    //Add Hidden Layers the biases of nets and the output weights in connect with output layer.
    [_krANN addHiddenLayerAtIndex:0 netBias:-0.4f netWeights:@[@-0.3f, @0.2f, @0.1f]]; //Net 5
    [_krANN addHiddenLayerAtIndex:0 netBias:0.2f netWeights:@[@-0.2f, @0.5f, @-0.1f]];  //Net 6

    [_krANN addHiddenLayerAtIndex:1 netBias:0.1f netWeights:@[@0.1f, @0.3f]];   //Net 7
    [_krANN addHiddenLayerAtIndex:1 netBias:0.25f netWeights:@[@0.2f, @0.1f]];  //Net 8
    [_krANN addHiddenLayerAtIndex:1 netBias:0.3f netWeights:@[@0.3f, @-0.4f]];  //Net 9

    [_krANN addHiddenLayerAtIndex:2 netBias:-0.25f netWeights:@[@0.4f, @0.3f]];  //Net 10
    [_krANN addHiddenLayerAtIndex:2 netBias:0.15f netWeights:@[@0.1f, @-0.2f]];  //Net 11

    //輸出層神經元偏權值, Net 12, Net 13
    //Add the output layer biases
    [_krANN addOutputBiases:@[@0.0f, @0.1f]];

    __block typeof(_krANN) _weakKRANN = _krANN;
    //訓練完成時( Training complete )
    [_krANN setTrainingCompletion:^(BOOL success, NSDictionary *trainedInfo, NSInteger totalTimes){
        if( success )
        {
            NSLog(@"Training done with total times : %i", totalTimes);
            NSLog(@"TrainedInfo 1 : %@", trainedInfo);

            //Start in checking the network is correctly trained.
            NSLog(@"======== Start in Verification ========");
            [_weakKRANN setTrainingCompletion:^(BOOL success, NSDictionary *trainedInfo, NSInteger totalTimes){
                NSLog(@"Training done with total times : %i", totalTimes);
                NSLog(@"TrainedInfo 2 : %@", trainedInfo);
            }];

            [_weakKRANN recoverNetwork];
            [_weakKRANN directOutputAtInputs:@[@1, @0.1, @0.5, @0.2]];
        }
    }];

    [_krANN training];
}

//Only setups patterns and output goals, and 1 output.
-(void)useSample2
{
    //Use f(x)=tanh(x)
    _krANN.activationFunction = KRANNActivationFunctionTanh;
    //How many hidden layers
    _krANN.hiddenLayerCount   = 0; //0 means let system decide the hidden layers number.

    [_krANN addPatterns:@[@1, @0.1, @0.5, @0.2] outputGoals:@[@0.7f]];    //Pattern 1, net 1, 2, 3, 4, and 1 output
    [_krANN addPatterns:@[@0, @-0.8, @0.3, @-0.9] outputGoals:@[@-0.1f]]; //Pattern 2, same as pattern 1
    [_krANN addPatterns:@[@1, @0.3, @0.1, @0.4] outputGoals:@[@0.9f]];    //Pattern 3, same as pattern 1

    __block typeof(_krANN) _weakKRANN = _krANN;
    //訓練完成時( Training complete )
    [_krANN setTrainingCompletion:^(BOOL success, NSDictionary *trainedInfo, NSInteger totalTimes){
        if( success )
        {
            NSLog(@"Training done with total times : %i", totalTimes);
            NSLog(@"TrainedInfo 1 : %@", trainedInfo);

            //Start in checking the network is correctly trained.
            NSLog(@"======== Start in Verification ========");
            [_weakKRANN setTrainingCompletion:^(BOOL success, NSDictionary *trainedInfo, NSInteger totalTimes){
                NSLog(@"Training done with total times : %i", totalTimes);
                NSLog(@"TrainedInfo 2 : %@", trainedInfo);
            }];

            [_weakKRANN recoverNetwork];
            [_weakKRANN directOutputAtInputs:@[@0, @-0.8, @0.3, @-0.9]];
        }
    }];

    [_krANN trainingByRandomWithSave];
}

//To learn and verify numbers 0 to 9. And only setups patterns and output goals, and 10 outputs.
-(void)useSample3
{
    _krANN.activationFunction = KRANNActivationFunctionSigmoid;
    //How many hidden layers, if you set 0 that means let system decide the hidden layers number.
    _krANN.hiddenLayerCount = 1;

    //1
    [_krANN addPatterns:@[@0, @0, @0, @0,
                          @0, @0, @0, @0,
                          @0, @0, @0, @0,
                          @0, @0, @0, @0,
                          @0, @0, @0, @0,
                          @0, @0, @0, @0,
                          @0, @0, @0, @1,
                          @1, @1, @1, @1,
                          @1, @1, @1, @1]
            outputGoals:@[@1, @0, @0, @0, @0, @0, @0, @0, @0, @0]];
    //2
    [_krANN addPatterns:@[@1, @0, @0, @0,
                          @1, @1, @1, @1,
                          @1, @1, @0, @0,
                          @0, @1, @0, @0,
                          @0, @1, @1, @0,
                          @0, @0, @1, @0,
                          @0, @0, @1, @1,
                          @1, @1, @1, @1,
                          @0, @0, @0, @1]
            outputGoals:@[@0, @1, @0, @0, @0, @0, @0, @0, @0, @0]];
    //3
    [_krANN addPatterns:@[@1, @0, @0, @0,
                          @1, @0, @0, @0,
                          @1, @1, @0, @0,
                          @0, @1, @0, @0,
                          @0, @1, @1, @0,
                          @0, @0, @1, @0,
                          @0, @0, @1, @1,
                          @1, @1, @1, @1,
                          @1, @1, @1, @1]
            outputGoals:@[@0, @0, @1, @0, @0, @0, @0, @0, @0, @0]];
    //4
    [_krANN addPatterns:@[@1, @1, @1, @1,
                          @1, @0, @0, @0,
                          @0, @0, @0, @0,
                          @0, @1, @0, @0,
                          @0, @0, @0, @0,
                          @0, @0, @1, @0,
                          @0, @0, @0, @1,
                          @1, @1, @1, @1,
                          @1, @1, @1, @1]
            outputGoals:@[@0, @0, @0, @1, @0, @0, @0, @0, @0, @0]];
    //5
    [_krANN addPatterns:@[@1, @1, @1, @1,
                          @1, @0, @0, @0,
                          @1, @1, @0, @0,
                          @0, @1, @0, @0,
                          @0, @1, @1, @0,
                          @0, @0, @1, @0,
                          @0, @0, @1, @1,
                          @0, @0, @0, @1,
                          @1, @1, @1, @1]
            outputGoals:@[@0, @0, @0, @0, @1, @0, @0, @0, @0, @0]];
    //6
    [_krANN addPatterns:@[@1, @1, @1, @1,
                          @1, @1, @1, @1,
                          @1, @1, @0, @0,
                          @0, @1, @0, @0,
                          @0, @1, @1, @0,
                          @0, @0, @1, @0,
                          @0, @0, @1, @1,
                          @0, @0, @0, @1,
                          @1, @1, @1, @1]
            outputGoals:@[@0, @0, @0, @0, @0, @1, @0, @0, @0, @0]];
    //7
    [_krANN addPatterns:@[@1, @0, @0, @0,
                          @0, @0, @0, @0,
                          @0, @1, @0, @0,
                          @0, @0, @0, @0,
                          @0, @0, @1, @0,
                          @0, @0, @0, @0,
                          @0, @0, @0, @1,
                          @1, @1, @1, @1,
                          @1, @1, @1, @1]
            outputGoals:@[@0, @0, @0, @0, @0, @0, @1, @0, @0, @0]];
    //8
    [_krANN addPatterns:@[@1, @1, @1, @1,
                          @1, @1, @1, @1,
                          @1, @1, @0, @0,
                          @0, @1, @0, @0,
                          @0, @1, @1, @0,
                          @0, @0, @1, @0,
                          @0, @0, @1, @1,
                          @1, @1, @1, @1,
                          @1, @1, @1, @1]
            outputGoals:@[@0, @0, @0, @0, @0, @0, @0, @1, @0, @0]];
    //9
    [_krANN addPatterns:@[@1, @1, @1, @1,
                          @1, @0, @0, @0,
                          @0, @1, @0, @0,
                          @0, @1, @0, @0,
                          @0, @0, @1, @0,
                          @0, @0, @1, @0,
                          @0, @0, @0, @1,
                          @1, @1, @1, @1,
                          @1, @1, @1, @1]
            outputGoals:@[@0, @0, @0, @0, @0, @0, @0, @0, @1, @0]];
    //0
    [_krANN addPatterns:@[@1, @1, @1, @1,
                          @1, @1, @1, @1,
                          @1, @1, @0, @0,
                          @0, @0, @0, @0,
                          @0, @1, @1, @0,
                          @0, @0, @0, @0,
                          @0, @0, @1, @1,
                          @1, @1, @1, @1,
                          @1, @1, @1, @1]
            outputGoals:@[@0, @0, @0, @0, @0, @0, @0, @0, @0, @1]];

    __block typeof(_krANN) _weakKRANN = _krANN;
    //訓練完成時( Training complete )
    [_krANN setTrainingCompletion:^(BOOL success, NSDictionary *trainedInfo, NSInteger totalTimes){
        if( success )
        {
            NSLog(@"Training done with total times : %i", totalTimes);
            NSLog(@"TrainedInfo 1 : %@", trainedInfo);

            //Start in checking the network is correctly trained.
            NSLog(@"======== Start in Verification ========");
            [_weakKRANN setTrainingCompletion:^(BOOL success, NSDictionary *trainedInfo, NSInteger totalTimes){
                NSLog(@"Training done with total times : %i", totalTimes);
                NSLog(@"TrainedInfo 2 : %@", trainedInfo);
            }];

            [_weakKRANN recoverNetwork];
            //Verified number " 7 ", and it has some defects.
            [_weakKRANN directOutputAtInputs:@[@1, @1, @1, @0,
                                               @0, @0, @0, @0,
                                               @0, @1, @0, @0,
                                               @0, @0, @0, @0,
                                               @0, @0, @1, @0,
                                               @0, @0, @0, @0,
                                               @0, @0, @0, @1,
                                               @1, @1, @1, @1,
                                               @1, @1, @1, @1]];

        }
    }];

    [_krANN trainingByRandomSettings];
}

- (void)viewDidLoad
{
    [super viewDidLoad];

    _krANN = [KRANN sharedNetwork];
    //_krANN.delegate = self;

    //學習速率
    _krANN.learningRate     = 0.8f;
    //收斂誤差值 ( Normally is 10^-3 or 10^-6 )
    _krANN.convergenceError = 0.000001f;
    //限制迭代次數
    _krANN.limitIteration   = 5000;

    //每一次的迭代( Per iteration-training )
    [_krANN setPerIteration:^(NSInteger times, NSDictionary *trainedInfo)
    {
        NSLog(@"Iteration times : %i", times);
        //NSLog(@"Iteration result : %f\n\n\n", [trainedInfo objectForKey:KRANNTrainedOutputResults]);
    }];

    [self useSample1];
    [self useSample2];
    [self useSample3];
}
@end

版本

V2.1.4

许可证

MIT。

备注

正在等待将性能提升和使用 EDBD（包括 QuickProp）方法来增强该算法。
关于用户指南，我没有时间在这里编写用户和技术指南。也许有一天我度假了很长时间，指南就会实施。