Poisson

Menghitung kerugian Poisson antara label dan prediksi.

loss = predictions - labels * log(predictions)

Penggunaan mandiri:

    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}, {0.f, 0.f} });
    Poisson poissonLoss = new Poisson(tf);
    Operand<TFloat32> result = poissonLoss.call(labels, predictions);
    // produces 0.5f
 

Memanggil dengan berat sampel:

    Operand<TFloat32> sampleWeight = tf.constant(new float[] {0.8f, 0.2f});
    Operand<TFloat32> result = poissonLoss.call(labels, predictions, sampleWeight);
    // produces 0.4f
 

Menggunakan tipe pengurangan SUM :

    Poisson poissonLoss = new Poisson(tf, Reduction.SUM);
    Operand<TFloat32> result = poissonLoss.call(labels, predictions);
    // produces 0.999f
 

Menggunakan tipe reduksi NONE :

    Poisson poissonLoss = new Poisson(tf, Reduction.NONE);
    Operand<TFloat32> result = poissonLoss.call(labels, predictions);
    // produces [0.999f, 0f]