Function block_inception_a

tests/utils/custom_layers/inception_blocks.py:19–67

Builds Inception-A block for Inception v4 network.

(inputs, scope=None, is_train=False)

def block_inception_a(inputs, scope=None, is_train=False):
    """Builds Inception-A block for Inception v4 network."""
    # By default use stride=1 and SAME padding

    with tf.variable_scope(name_or_scope=scope, default_name='BlockInceptionA', values=[inputs]):
        with tf.variable_scope('Branch_0'):
            branch_0, _ = conv_module(
                inputs, n_out_channel=96, filter_size=(1, 1), strides=(1, 1), padding='SAME', batch_norm_init=None,
                is_train=is_train, use_batchnorm=True, activation_fn='ReLU', name='Conv2d_0a_1x1'
            )

        with tf.variable_scope('Branch_1'):
            branch_1, _ = conv_module(
                inputs, n_out_channel=64, filter_size=(1, 1), strides=(1, 1), padding='SAME', batch_norm_init=None,
                is_train=is_train, use_batchnorm=True, activation_fn='ReLU', name='Conv2d_0a_1x1'
            )

            branch_1, _ = conv_module(
                branch_1, n_out_channel=96, filter_size=(3, 3), strides=(1, 1), padding='SAME', batch_norm_init=None,
                is_train=is_train, use_batchnorm=True, activation_fn='ReLU', name='Conv2d_0b_3x3'
            )

        with tf.variable_scope('Branch_2'):
            branch_2, _ = conv_module(
                inputs, n_out_channel=64, filter_size=(1, 1), strides=(1, 1), padding='SAME', batch_norm_init=None,
                is_train=is_train, use_batchnorm=True, activation_fn='ReLU', name='Conv2d_0a_1x1'
            )

            branch_2, _ = conv_module(
                branch_2, n_out_channel=96, filter_size=(3, 3), strides=(1, 1), padding='SAME', batch_norm_init=None,
                is_train=is_train, use_batchnorm=True, activation_fn='ReLU', name='Conv2d_0b_3x3'
            )

            branch_2, _ = conv_module(
                branch_2, n_out_channel=96, filter_size=(3, 3), strides=(1, 1), padding='SAME', batch_norm_init=None,
                is_train=is_train, use_batchnorm=True, activation_fn='ReLU', name='Conv2d_0c_3x3'
            )

        with tf.variable_scope('Branch_3'):
            branch_3 = tl.layers.MeanPool2d(
                inputs, filter_size=(3, 3), strides=(1, 1), padding='SAME', name='AvgPool_0a_3x3'
            )

            branch_3, _ = conv_module(
                branch_3, n_out_channel=96, filter_size=(1, 1), strides=(1, 1), padding='SAME', batch_norm_init=None,
                is_train=is_train, use_batchnorm=True, activation_fn='ReLU', name='Conv2d_0b_1x1'
            )

        return tl.layers.ConcatLayer([branch_0, branch_1, branch_2, branch_3], concat_dim=3, name='concat_layer')


def block_reduction_a(inputs, scope=None, is_train=False):