dl4j issues

gistfile1.txt

Exception in thread "main" java.lang.IllegalArgumentException: Shape must be <= buffer length
	at org.nd4j.linalg.api.ndarray.BaseNDArray.<init>(BaseNDArray.java:133)
	at org.nd4j.linalg.jblas.NDArray.<init>(NDArray.java:65)
	at org.nd4j.linalg.jblas.JblasNDArrayFactory.create(JblasNDArrayFactory.java:229)
	at org.nd4j.linalg.factory.Nd4j.create(Nd4j.java:2919)
	at org.nd4j.linalg.api.ndarray.BaseNDArray.create(BaseNDArray.java:3305)
	at org.nd4j.linalg.api.ndarray.BaseNDArray.reshape(BaseNDArray.java:3348)
	at org.nd4j.linalg.api.ndarray.BaseNDArray.reshape(BaseNDArray.java:3372)
	at org.nd4j.linalg.api.ndarray.BaseNDArray.reshape(BaseNDArray.java:3605)
	at org.deeplearning4j.nn.conf.preprocessor.RnnToFeedForwardPreProcessor.preProcess(RnnToFeedForwardPreProcessor.java:42)
	at org.deeplearning4j.nn.multilayer.MultiLayerNetwork.activationFromPrevLayer(MultiLayerNetwork.java:450)
	at org.deeplearning4j.nn.multilayer.MultiLayerNetwork.feedForward(MultiLayerNetwork.java:513)
	at org.deeplearning4j.nn.multilayer.MultiLayerNetwork.feedForward(MultiLayerNetwork.java:499)
	at org.deeplearning4j.nn.multilayer.MultiLayerNetwork.output(MultiLayerNetwork.java:1250)
	at org.deeplearning4j.nn.multilayer.MultiLayerNetwork.output(MultiLayerNetwork.java:1266)
	at org.deeplearning4j.nn.multilayer.MultiLayerNetwork.predict(MultiLayerNetwork.java:1143)
	at testicle.LSTMTest.main(LSTMTest.java:130)
	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
	at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
	at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
	at java.lang.reflect.Method.invoke(Method.java:483)
	at com.intellij.rt.execution.application.AppMain.main(AppMain.java:140)
	
	

gistfile2.txt

	public static void main(String... args)throws Exception {

        /*
        final int numRows = 1;
        final int numColumns = 1;
        int outputNum = 1;
        int numSamples = 10;
        int batchSize = 10;
        int iterations = 1;
        int seed = 123;
        int listenerFreq = iterations/5;
        //log.info("Loading data...");
        //MnistDataFetcher fetcher = new MnistDataFetcher(true);
        log.info("Building model...");
        NeuralNetConfiguration conf = new NeuralNetConfiguration.Builder()
                .layer(new LSTM())
                .nIn(numRows * numColumns)
                .nOut(numRows * numColumns)
                .activationFunction("sigmoid")
                .optimizationAlgo(OptimizationAlgorithm.LBFGS)
                .constrainGradientToUnitNorm(true)
                .lossFunction(LossFunctions.LossFunction.RMSE_XENT)
                .build();
        Layer model = LayerFactories.getFactory(conf.getLayer()).create(conf);
        model.setListeners(Arrays.asList((IterationListener) new ScoreIterationListener(listenerFreq)));
        log.info("Training model...");
       // for(int i=0 ; i < (numSamples/batchSize); i++) {
            DataSet mnist = getDataSet();
            model.fit(mnist.getFeatureMatrix());
        //}
// Generative model - unsupervised and its time series based which requires different evaluation technique

        */

        Nd4j.getRandom().setSeed(12345L);
        int timeSeriesLength = 2;
        int nIn = 1;
        int layerSize = 2;
        int nOut = 1;
        int miniBatchSize = 11;
        MultiLayerConfiguration conf = new NeuralNetConfiguration.Builder()
                .weightInit(WeightInit.DISTRIBUTION).dist(new NormalDistribution(0,0.1))
                .regularization(false)
                .updater(Updater.NONE)
                .seed(12345L)
                .list(2)
                .layer(0, new GravesLSTM.Builder().nIn(nIn).nOut(layerSize).activation("tanh").build())
                .layer(1, new OutputLayer.Builder(LossFunction.MCXENT).activation("softmax").nIn(layerSize).nOut(nOut).build())
                .inputPreProcessor(1, new RnnToFeedForwardPreProcessor(timeSeriesLength))
                .pretrain(false).backprop(true)
                .build();
        MultiLayerNetwork mln = new MultiLayerNetwork(conf);
        mln.init();
        //huutaa niin vitusti jtn helper palikkaa tahan
        Random r = new Random(12345L);
        INDArray input = Nd4j.zeros(miniBatchSize,nIn,timeSeriesLength);
        for( int i=0; i<miniBatchSize; i++ ){
            for( int j=0; j<nIn; j++ ){
                for( int k=0; k<timeSeriesLength; k++ ){
                    input.putScalar(new int[]{i,j,k},r.nextDouble()-0.5);
                }
            }
        }
        INDArray labels = Nd4j.zeros(miniBatchSize*timeSeriesLength,nOut); //Would be this shape after reshaping 3d -> 2d for output layer
        for( int i=0; i<labels.size(0); i++){
            int idx = r.nextInt(nOut);
            labels.putScalar(new int[]{i,idx}, 1.0f);
        }

        //DataSet ds = new DataSet(input,labels);
        mln.fit(input,labels);

        log.info("los tulos: {}",mln.predict(Nd4j.zeros(1,nIn)));

    }