Keras で独自の損失関数を実装するにはどうすればよいですか?

dice coefficient = (2 * intersection) / (sum(ground_truth) + sum(predictions))

<code class="python">import keras.backend as K

def dice_coef(y_true, y_pred, smooth, thresh):
    y_pred = y_pred > thresh
    y_true_f = K.flatten(y_true)
    y_pred_f = K.flatten(y_pred)
    intersection = K.sum(y_true_f * y_pred_f)

    return (2. * intersection + smooth) / (K.sum(y_true_f) + K.sum(y_pred_f) + smooth)</code>

<code class="python">def dice_loss(smooth, thresh):
  def dice(y_true, y_pred):
    return -dice_coef(y_true, y_pred, smooth, thresh)
  return dice</code>

<code class="python"># build model
model = my_model()

# get the loss function
model_dice = dice_loss(smooth=1e-5, thresh=0.5)

# compile model
model.compile(loss=model_dice)</code>