c3ac58c90f2fe330eb036744fba3af2c14426fdc,doc_seg/model.py,,model_fn,#Any#Any#Any#Any#,9

Before Change

        if training_params.training_margin > 0:
            margin = training_params.training_margin
            loss = tf.reduce_mean(per_pixel_loss[:, margin:-margin, margin:-margin], name="loss")
        else:
            loss = tf.reduce_mean(per_pixel_loss, name="loss")

        loss += regularized_loss

After Change

            def _fn(_in):
                output, shape = _in
                return tf.reduce_mean(output[margin:shape[0]-margin, margin:shape[1]-margin])
            per_img_loss = tf.map_fn(_fn, (per_pixel_loss, input_shapes), dtype=tf.float32)
            loss = tf.reduce_mean(per_img_loss, name="loss")

        loss += regularized_loss
    else:

Instances