MeanSquaredLogarithmicError

Calcula os erros logarítmicos médios quadrados entre rótulos e previsões.

loss = square(log(labels + 1.) - log(predictions + 1.))

Uso autônomo:

    Operand<TFloat32> labels =
        tf.constant(new float[][] { {0.f, 1.f}, {0.f, 0.f} });
    Operand<TFloat32> predictions =
        tf.constant(new float[][] { {1.f, 1.f}, {1.f, 0.f} });
    MeanSquaredLogarithmicError msle = new MeanSquaredLogarithmicError(tf);
    Operand<TFloat32> result = msle.call(labels, predictions);
    // produces 0.240f
 

Chamando com peso de amostra:

    Operand<TFloat32> sampleWeight = tf.constant(new float[] {0.7f, 0.3f});
    Operand<TFloat32> result = msle.call(labels, predictions, sampleWeight);
    // produces 0.120f
 

Usando o tipo de redução SUM :

    MeanSquaredLogarithmicError msle = new MeanSquaredLogarithmicError(tf, Reduction.SUM);
    Operand<TFloat32> result = msle.call(labels, predictions);
    // produces 0.480f
 

Usando NONE tipo de redução:

    MeanSquaredLogarithmicError msle = new MeanSquaredLogarithmicError(tf, Reduction.NONE);
    Operand<TFloat32> result = msle.call(labels, predictions);
    // produces [0.240f, 0.240f]