conv0 = tf.compat.v1.layers.Conv2D(
     filters=filters,
     kernel_size=kernel_size,
     strides=2,
     padding=('SAME' if strides == 1 else 'VALID'),
     use_bias=False,
     kernel_initializer=tf.variance_scaling_initializer(),
     data_format=data_format)

    # Use the image size without space-to-depth transform as the input of conv0.
    batch_size, h, w, channel = inputs.get_shape().as_list()
    conv0.build([

Computes scaled exponential linear: `scale * alpha * (exp(features) - 1)`
  }];

  let description = [{
if < 0, `scale * features` otherwise.

To be used together with
`initializer = tf.variance_scaling_initializer(factor=1.0, mode='FAN_IN')`.
For correct dropout, use `tf.contrib.nn.alpha_dropout`.

See [Self-Normalizing Neural Networks](https://arxiv.org/abs/1706.02515)
  }];

  let arguments = (ins