9f66664f055e7943d157ba94b7bd5b0bc5b3d54e,models/experimental/deeplab/model.py,,loss_fn,#Any#Any#Any#Any#,30

Before Change

      params["image_pyramid"],
      weight_decay=params["weight_decay"],
      is_training=mode == tf.estimator.ModeKeys.TRAIN,
      fine_tune_batch_norm=params["fine_tune_batch_norm"]
  )

  for output, num_classes in params["outputs_to_num_classes"].items():
    add_softmax_cross_entropy_loss_for_each_scale(

After Change

        upsample_logits=params["upsample_logits"],
        scope=output)

  losses = tf.add_n(tf.losses.get_losses())
  l2_loss = []
  for v in tf.trainable_variables():
    if "BatchNorm" not in v.name and "weights" in v.name:
      l2_loss.append(tf.nn.l2_loss(v))
  loss = losses + params["weight_decay"] * tf.add_n(l2_loss)
  return loss


def _create_eval_metric(features, labels, params):

Instances