Examples with TransformationLayerArchitecture org.apache.ignite.ml.nn.architecture.TransformationLayerArchitecture used on opensource projects

Example 1 with TransformationLayerArchitecture

use of org.apache.ignite.ml.nn.architecture.TransformationLayerArchitecture in project ignite by apache.

the class MultilayerPerceptron method initLayers.

/**
 * Init layers parameters with initializer.
 *
 * @param initializer Parameters initializer.
 */
private void initLayers(MLPInitializer initializer) {
    int prevSize = architecture.inputSize();
    for (int i = 1; i < architecture.layersCount(); i++) {
        TransformationLayerArchitecture layerCfg = architecture.transformationLayerArchitecture(i);
        int neuronsCnt = layerCfg.neuronsCount();
        DenseLocalOnHeapMatrix weights = new DenseLocalOnHeapMatrix(neuronsCnt, prevSize);
        initializer.initWeights(weights);
        DenseLocalOnHeapVector biases = null;
        if (layerCfg.hasBias()) {
            biases = new DenseLocalOnHeapVector(neuronsCnt);
            initializer.initBiases(biases);
        }
        layers.add(new MLPLayer(weights, biases));
        prevSize = layerCfg.neuronsCount();
    }
}

Example 2 with TransformationLayerArchitecture

use of org.apache.ignite.ml.nn.architecture.TransformationLayerArchitecture in project ignite by apache.

the class MultilayerPerceptron method forwardPass.

/**
 * Perform forward pass and if needed write state of outputs of each layer.
 *
 * @param val Value to perform computation on.
 * @param state State object to write state into.
 * @param writeState Flag indicating need to write state.
 */
public Matrix forwardPass(Matrix val, MLPState state, boolean writeState) {
    Matrix res = val;
    if (below != null)
        res = below.forwardPass(val, state, writeState);
    for (int i = 1; i < architecture.layersCount(); i++) {
        MLPLayer curLayer = layers.get(i - 1);
        res = curLayer.weights.times(res);
        TransformationLayerArchitecture layerCfg = this.architecture.transformationLayerArchitecture(i);
        if (layerCfg.hasBias()) {
            ReplicatedVectorMatrix biasesMatrix = new ReplicatedVectorMatrix(biases(i), res.columnSize(), true);
            res = res.plus(biasesMatrix);
        }
        state.linearOutput.add(res);
        // If we do not write state, we can overwrite result.
        if (writeState)
            res = res.copy();
        res = res.map(layerCfg.activationFunction());
        state.activatorsOutput.add(res);
    }
    return res;
}