Python keras.layers.add() Examples

def test_tiny_conv_dilated(self, model_precision=_MLMODEL_FULL_PRECISION):
        np.random.seed(1988)
        input_dim = 10
        input_shape = (input_dim, input_dim, 1)
        num_kernels, kernel_height, kernel_width = 3, 5, 5

        # Define a model
        model = Sequential()
        model.add(
            Conv2D(
                input_shape=input_shape,
                dilation_rate=(2, 2),
                filters=num_kernels,
                kernel_size=(kernel_height, kernel_width),
            )
        )

        # Set some random weights
        model.set_weights([np.random.rand(*w.shape) for w in model.get_weights()])

        # Test the keras model
        self._test_model(model, model_precision=model_precision) 

def resnet_first_block_first_module(input, channel_depth):
    residual_input = input
    stride = 1

    residual_input = Conv2D(channel_depth, kernel_size=1, strides=1, padding="same", kernel_initializer="he_normal")(
        residual_input)
    residual_input = BatchNormalization()(residual_input)

    input = Conv2D(int(channel_depth / 4), kernel_size=1, strides=stride, padding="same",
                   kernel_initializer="he_normal")(input)
    input = BatchNormalization()(input)
    input = Activation("relu")(input)

    input = Conv2D(int(channel_depth / 4), kernel_size=3, strides=stride, padding="same",
                   kernel_initializer="he_normal")(input)
    input = BatchNormalization()(input)
    input = Activation("relu")(input)

    input = Conv2D(channel_depth, kernel_size=1, strides=stride, padding="same", kernel_initializer="he_normal")(input)
    input = BatchNormalization()(input)

    input = add([input, residual_input])
    input = Activation("relu")(input)

    return input 

def _octresnet_final_bottleneck_block(ip, filters, alpha=0.5, strides=(1, 1),
                                      downsample_shortcut=False,
                                      expansion=4):

    x_high_res, x_low_res = ip

    x_high, x_low = oct_conv_bn_relu(x_high_res, x_low_res, filters, kernel_size=(1, 1),
                                     alpha=alpha)

    x_high, x_low = oct_conv_bn_relu(x_high, x_low, filters, kernel_size=(3, 3),
                                     strides=strides, alpha=alpha)

    final_filters = int(filters * expansion)
    x_high = final_oct_conv_bn_relu(x_high, x_low, final_filters, kernel_size=(1, 1),
                                    activation=False)

    if downsample_shortcut:
        x_high_res = final_oct_conv_bn_relu(x_high_res, x_low_res, final_filters, kernel_size=(1, 1),
                                            strides=strides, activation=False)

    x = add([x_high, x_high_res])
    x = ReLU()(x)

    return x 

def _bottleneck_original(ip, filters, strides=(1, 1), downsample_shortcut=False,
                         expansion=4):

    final_filters = int(filters * expansion)

    shortcut = ip

    x = _conv_bn_relu(ip, filters, kernel_size=(1, 1))
    x = _conv_bn_relu(x, filters, kernel_size=(3, 3), strides=strides)
    x = _conv_bn_relu(x, final_filters, kernel_size=(1, 1), activation=False)

    if downsample_shortcut:
        shortcut = _conv_block(shortcut, final_filters, kernel_size=(1, 1),
                               strides=strides)

    x = add([x, shortcut])
    x = ReLU()(x)

    return x 

def test_tiny_inner_product(self, model_precision=_MLMODEL_FULL_PRECISION):
        np.random.seed(1988)

        # Define a model
        model = Sequential()
        model.add(Dense(2, input_shape=(2,)))

        # Test all zeros
        model.set_weights([np.random.rand(*w.shape) for w in model.get_weights()])
        self._test_model(model, mode="zeros", model_precision=model_precision)

        # Test all ones
        model.set_weights([np.ones(w.shape) for w in model.get_weights()])
        self._test_model(model, mode="ones", model_precision=model_precision)

        # Test random
        model.set_weights([np.random.rand(*w.shape) for w in model.get_weights()])
        self._test_model(model, model_precision=model_precision) 

def channel_spatial_squeeze_excite(input, ratio=16):
    ''' Create a spatial squeeze-excite block

    Args:
        input: input tensor
        filters: number of output filters

    Returns: a keras tensor

    References
    -   [Squeeze and Excitation Networks](https://arxiv.org/abs/1709.01507)
    -   [Concurrent Spatial and Channel Squeeze & Excitation in Fully Convolutional Networks](https://arxiv.org/abs/1803.02579)
    '''

    cse = squeeze_excite_block(input, ratio)
    sse = spatial_squeeze_excite_block(input)

    x = add([cse, sse])
    return x 

def channel_spatial_squeeze_excite(input_tensor, ratio=16):
    """ Create a spatial squeeze-excite block

    Args:
        input_tensor: input Keras tensor
        ratio: number of output filters

    Returns: a Keras tensor

    References
    -   [Squeeze and Excitation Networks](https://arxiv.org/abs/1709.01507)
    -   [Concurrent Spatial and Channel Squeeze & Excitation in Fully Convolutional Networks](https://arxiv.org/abs/1803.02579)
    """

    cse = squeeze_excite_block(input_tensor, ratio)
    sse = spatial_squeeze_excite_block(input_tensor)

    x = add([cse, sse])
    return x 

def test_tiny_conv_ones(self, model_precision=_MLMODEL_FULL_PRECISION):
        np.random.seed(1988)
        input_dim = 10
        input_shape = (input_dim, input_dim, 1)
        num_kernels, kernel_height, kernel_width = 3, 5, 5

        # Define a model
        model = Sequential()
        model.add(
            Conv2D(
                input_shape=input_shape,
                filters=num_kernels,
                kernel_size=(kernel_height, kernel_width),
            )
        )

        # Set some random weights
        model.set_weights([np.ones(w.shape) for w in model.get_weights()])

        # Test the keras model
        self._test_model(model, model_precision=model_precision) 

def d_block(inp, fil, p = True):

    skip = Conv2D(fil, 1, padding = 'same', kernel_initializer = 'he_normal')(inp)

    out = Conv2D(filters = fil, kernel_size = 3, padding = 'same', kernel_initializer = 'he_normal')(inp)
    out = LeakyReLU(0.2)(out)

    out = Conv2D(filters = fil, kernel_size = 3, padding = 'same', kernel_initializer = 'he_normal')(out)
    out = LeakyReLU(0.2)(out)

    out = Conv2D(fil, 1, padding = 'same', kernel_initializer = 'he_normal')(out)

    out = add([out, skip])
    out = LeakyReLU(0.2)(out)

    if p:
        out = AveragePooling2D()(out)

    return out 

def g_block(inp, fil, u = True):

    if u:
        out = UpSampling2D(interpolation = 'bilinear')(inp)
    else:
        out = Activation('linear')(inp)

    skip = Conv2D(fil, 1, padding = 'same', kernel_initializer = 'he_normal')(out)

    out = Conv2D(filters = fil, kernel_size = 3, padding = 'same', kernel_initializer = 'he_normal')(out)
    out = LeakyReLU(0.2)(out)

    out = Conv2D(filters = fil, kernel_size = 3, padding = 'same', kernel_initializer = 'he_normal')(out)
    out = LeakyReLU(0.2)(out)

    out = Conv2D(fil, 1, padding = 'same', kernel_initializer = 'he_normal')(out)

    out = add([out, skip])
    out = LeakyReLU(0.2)(out)

    return out 

def test_tiny_conv_random(self, model_precision=_MLMODEL_FULL_PRECISION):
        np.random.seed(1988)
        input_dim = 10
        input_shape = (input_dim, input_dim, 1)
        num_kernels, kernel_height, kernel_width = 3, 5, 5

        # Define a model
        model = Sequential()
        model.add(
            Conv2D(
                input_shape=input_shape,
                filters=num_kernels,
                kernel_size=(kernel_height, kernel_width),
            )
        )

        # Set some random weights
        model.set_weights([np.random.rand(*w.shape) for w in model.get_weights()])

        # Test the keras model
        self._test_model(model, model_precision=model_precision) 

def SepConv_BN(x, filters, prefix, stride=1, kernel_size=3, rate=1, depth_activation=False, epsilon=1e-3):
    """ SepConv with BN between depthwise & pointwise. Optionally add activation after BN
        Implements right "same" padding for even kernel sizes
        Args:
            x: input tensor
            filters: num of filters in pointwise convolution
            prefix: prefix before name
            stride: stride at depthwise conv
            kernel_size: kernel size for depthwise convolution
            rate: atrous rate for depthwise convolution
            depth_activation: flag to use activation between depthwise & poinwise convs
            epsilon: epsilon to use in BN layer
    """

    if stride == 1:
        depth_padding = 'same'
    else:
        kernel_size_effective = kernel_size + (kernel_size - 1) * (rate - 1)
        pad_total = kernel_size_effective - 1
        pad_beg = pad_total // 2
        pad_end = pad_total - pad_beg
        x = ZeroPadding2D((pad_beg, pad_end))(x)
        depth_padding = 'valid'

    if not depth_activation:
        x = Activation('relu')(x)
    x = DepthwiseConv2D((kernel_size, kernel_size), strides=(stride, stride), dilation_rate=(rate, rate),
                        padding=depth_padding, use_bias=False, name=prefix + '_depthwise')(x)
    x = BatchNormalization(name=prefix + '_depthwise_BN', epsilon=epsilon)(x)
    if depth_activation:
        x = Activation('relu')(x)
    x = Conv2D(filters, (1, 1), padding='same',
               use_bias=False, name=prefix + '_pointwise')(x)
    x = BatchNormalization(name=prefix + '_pointwise_BN', epsilon=epsilon)(x)
    if depth_activation:
        x = Activation('relu')(x)

    return x 

def _inverted_res_block(inputs, expansion, stride, alpha, filters, block_id, skip_connection, rate=1):
    in_channels = inputs._keras_shape[-1]
    pointwise_conv_filters = int(filters * alpha)
    pointwise_filters = _make_divisible(pointwise_conv_filters, 8)
    x = inputs
    prefix = 'expanded_conv_{}_'.format(block_id)
    if block_id:
        # Expand

        x = Conv2D(expansion * in_channels, kernel_size=1, padding='same',
                   use_bias=False, activation=None,
                   name=prefix + 'expand')(x)
        x = BatchNormalization(epsilon=1e-3, momentum=0.999,
                               name=prefix + 'expand_BN')(x)
        x = Activation(relu6, name=prefix + 'expand_relu')(x)
    else:
        prefix = 'expanded_conv_'
    # Depthwise
    x = DepthwiseConv2D(kernel_size=3, strides=stride, activation=None,
                        use_bias=False, padding='same', dilation_rate=(rate, rate),
                        name=prefix + 'depthwise')(x)
    x = BatchNormalization(epsilon=1e-3, momentum=0.999,
                           name=prefix + 'depthwise_BN')(x)

    x = Activation(relu6, name=prefix + 'depthwise_relu')(x)

    # Project
    x = Conv2D(pointwise_filters,
               kernel_size=1, padding='same', use_bias=False, activation=None,
               name=prefix + 'project')(x)
    x = BatchNormalization(epsilon=1e-3, momentum=0.999,
                           name=prefix + 'project_BN')(x)

    if skip_connection:
        return Add(name=prefix + 'add')([inputs, x])

    # if in_channels == pointwise_filters and stride == 1:
    #    return Add(name='res_connect_' + str(block_id))([inputs, x])

    return x 

def test_inner_product_random(self, model_precision=_MLMODEL_FULL_PRECISION):
        np.random.seed(1988)

        # Define a model
        model = Sequential()
        model.add(Dense(1000, input_shape=(100,)))

        # Set some random weights
        model.set_weights([np.random.rand(*w.shape) for w in model.get_weights()])

        # Test the keras model
        self._test_model(model, model_precision=model_precision) 

def _resnet_bottleneck_block(input, filters, k=1, strides=(1, 1)):
    ''' Adds a pre-activation resnet block with bottleneck layers

    Args:
        input: input tensor
        filters: number of output filters
        k: width factor
        strides: strides of the convolution layer

    Returns: a keras tensor
    '''
    init = input
    channel_axis = 1 if K.image_data_format() == "channels_first" else -1
    bottleneck_expand = 4

    x = BatchNormalization(axis=channel_axis)(input)
    x = Activation('relu')(x)

    if strides != (1, 1) or init._keras_shape[channel_axis] != bottleneck_expand * filters * k:
        init = Conv2D(bottleneck_expand * filters * k, (1, 1), padding='same', kernel_initializer='he_normal',
                      use_bias=False, strides=strides)(x)

    x = Conv2D(filters * k, (1, 1), padding='same', kernel_initializer='he_normal',
               use_bias=False)(x)
    x = BatchNormalization(axis=channel_axis)(x)
    x = Activation('relu')(x)

    x = Conv2D(filters * k, (3, 3), padding='same', kernel_initializer='he_normal',
               use_bias=False, strides=strides)(x)
    x = BatchNormalization(axis=channel_axis)(x)
    x = Activation('relu')(x)

    x = Conv2D(bottleneck_expand * filters * k, (1, 1), padding='same', kernel_initializer='he_normal',
               use_bias=False)(x)

    # global context block
    x = global_context_block(x)

    m = add([x, init])
    return m 

def test_dense_softmax(self):
        np.random.seed(1988)

        # Define a model
        model = Sequential()
        model.add(Dense(32, input_shape=(32,), activation="softmax"))

        # Set some random weights
        model.set_weights([np.random.rand(*w.shape) for w in model.get_weights()])

        # Test the keras model
        self._test_model(model) 

def test_dense_elu(self):
        np.random.seed(1988)

        # Define a model
        model = Sequential()
        model.add(Dense(32, input_shape=(32,), activation="elu"))

        # Set some random weights
        model.set_weights([np.random.rand(*w.shape) for w in model.get_weights()])

        # Test the keras model
        self._test_model(model) 

def test_dense_selu(self):
        np.random.seed(1988)

        # Define a model
        model = Sequential()
        model.add(Dense(32, input_shape=(32,), activation="selu"))

        # Set some random weights
        model.set_weights([np.random.rand(*w.shape) for w in model.get_weights()])

        # Test the keras model
        self._test_model(model) 

def test_tiny_conv_random_input_shape_dict(
        self, model_precision=_MLMODEL_FULL_PRECISION
    ):
        np.random.seed(1988)
        H, W, C = 10, 20, 5
        input_shape = (None, H, W, C)
        num_kernels, kernel_height, kernel_width = 3, 5, 5

        # Define a model
        model = Sequential()
        model.add(
            Conv2D(
                input_shape=(None, None, C),
                filters=num_kernels,
                kernel_size=(kernel_height, kernel_width),
            )
        )

        # Set some random weights
        model.set_weights([np.random.rand(*w.shape) for w in model.get_weights()])

        # Test the keras model
        self._test_model(
            model,
            input_name_shape_dict={"data": input_shape},
            model_precision=model_precision,
        ) 

def Rk(self, x0):
        k = int(x0.shape[-1])
        # first layer
        x = ReflectionPadding2D((1,1))(x0)
        x = Conv2D(filters=k, kernel_size=3, strides=1, padding='valid')(x)
        x = self.normalization(axis=3, center=True, epsilon=1e-5)(x, training=True)
        x = Activation('relu')(x)
        # second layer
        x = ReflectionPadding2D((1, 1))(x)
        x = Conv2D(filters=k, kernel_size=3, strides=1, padding='valid')(x)
        x = self.normalization(axis=3, center=True, epsilon=1e-5)(x, training=True)
        # merge
        x = add([x, x0])
        return x 

def _xception_block(inputs, depth_list, prefix, skip_connection_type, stride,
                    rate=1, depth_activation=False, return_skip=False):
    """ Basic building block of modified Xception network
        Args:
            inputs: input tensor
            depth_list: number of filters in each SepConv layer. len(depth_list) == 3
            prefix: prefix before name
            skip_connection_type: one of {'conv','sum','none'}
            stride: stride at last depthwise conv
            rate: atrous rate for depthwise convolution
            depth_activation: flag to use activation between depthwise & pointwise convs
            return_skip: flag to return additional tensor after 2 SepConvs for decoder
            """
    residual = inputs
    for i in range(3):
        residual = SepConv_BN(residual,
                              depth_list[i],
                              prefix + '_separable_conv{}'.format(i + 1),
                              stride=stride if i == 2 else 1,
                              rate=rate,
                              depth_activation=depth_activation)
        if i == 1:
            skip = residual
    if skip_connection_type == 'conv':
        shortcut = _conv2d_same(inputs, depth_list[-1], prefix + '_shortcut',
                                kernel_size=1,
                                stride=stride)
        shortcut = BatchNormalization(name=prefix + '_shortcut_BN')(shortcut)
        outputs = layers.add([residual, shortcut])
    elif skip_connection_type == 'sum':
        outputs = layers.add([residual, inputs])
    elif skip_connection_type == 'none':
        outputs = residual
    if return_skip:
        return outputs, skip
    else:
        return outputs 

def identity_block(input_tensor, kernel_size, filters, stage, block):
    """The identity block is the block that has no conv layer at shortcut.

    # Arguments
        input_tensor: input tensor
        kernel_size: default 3, the kernel size of middle conv layer at main path
        filters: list of integers, the filterss of 3 conv layer at main path
        stage: integer, current stage label, used for generating layer names
        block: 'a','b'..., current block label, used for generating layer names

    # Returns
        Output tensor for the block.
    """
    filters1, filters2, filters3 = filters
    if K.image_data_format() == 'channels_last':
        bn_axis = 3
    else:
        bn_axis = 1
    conv_name_base = 'res' + str(stage) + block + '_branch'
    bn_name_base = 'bn' + str(stage) + block + '_branch'

    x = Conv2D(filters1, (1, 1), name=conv_name_base + '2a')(input_tensor)
    x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2a')(x)
    x = Activation('relu')(x)

    x = Conv2D(filters2, kernel_size,
               padding='same', name=conv_name_base + '2b')(x)
    x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2b')(x)
    x = Activation('relu')(x)

    x = Conv2D(filters3, (1, 1), name=conv_name_base + '2c')(x)
    x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2c')(x)

    x = layers.add([x, input_tensor])
    x = Activation('relu')(x)
    return x 

def identity_block(input_tensor, kernel_size, filters, stage, block):
    """The identity block is the block that has no conv layer at shortcut.
    # Arguments
        input_tensor: input tensor
        kernel_size: defualt 3, the kernel size of middle conv layer at
                     main path
        filters: list of integers, the filterss of 3 conv layer at main path
        stage: integer, current stage label, used for generating layer names
        block: 'a','b'..., current block label, used for generating layer names
    # Returns
        Output tensor for the block.
    """
    filters1, filters2, filters3 = filters

    if IMAGE_ORDERING == 'channels_last':
        bn_axis = 3
    else:
        bn_axis = 1

    conv_name_base = 'res' + str(stage) + block + '_branch'
    bn_name_base = 'bn' + str(stage) + block + '_branch'

    x = Conv2D(filters1, (1, 1), data_format=IMAGE_ORDERING,
               name=conv_name_base + '2a')(input_tensor)
    x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2a')(x)
    x = Activation('relu')(x)

    x = Conv2D(filters2, kernel_size, data_format=IMAGE_ORDERING,
               padding='same', name=conv_name_base + '2b')(x)
    x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2b')(x)
    x = Activation('relu')(x)

    x = Conv2D(filters3, (1, 1), data_format=IMAGE_ORDERING,
               name=conv_name_base + '2c')(x)
    x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2c')(x)

    x = layers.add([x, input_tensor])
    x = Activation('relu')(x)
    return x 

def build_REDNet(nb_layers, input_size, nb_filters=32, k_size=3, dropout=0, strides=1, every=1):
    # -> CONV/FC -> BatchNorm -> ReLu(or other activation) -> Dropout -> CONV/FC ->  # https://arxiv.org/pdf/1502.03167.pdf
    input_img = Input(shape=(input_size, input_size, 1))
    x = input_img

    if K.image_data_format() == 'channels_last':
        bn_axis = 3
    else:
        bn_axis = 1

    encoderLayers = [None] * nb_layers

    for i in range(nb_layers):
        x = Conv2D(nb_filters, kernel_size=k_size, strides=strides, padding='same')(x)
        x = BatchNormalization(axis=bn_axis)(x)
        x = Activation('relu')(x)
        if dropout > 0:
            x = Dropout(dropout)(x)
        encoderLayers[i] = x

    encoded = x

    for i in range(nb_layers):
        ind = nb_layers - i - 1
        x = layers.add([x, encoderLayers[ind]])

        x = Conv2DTranspose(nb_filters, kernel_size=k_size, strides=strides, padding='same')(x)
        x = BatchNormalization(axis=bn_axis)(x)
        x = Activation('relu')(x)
        if dropout > 0:
            x = Dropout(dropout)(x)

    decoded = Conv2D(1, kernel_size=k_size, strides=1, padding='same', activation='sigmoid')(x)

    autoencoder = Model(input_img, decoded)

    return autoencoder, encoded, decoded 

def _inverted_res_block(inputs, expansion, stride, alpha, filters, block_id, skip_connection, rate=1):
    in_channels = inputs._keras_shape[-1]
    pointwise_conv_filters = int(filters * alpha)
    pointwise_filters = _make_divisible(pointwise_conv_filters, 8)
    x = inputs
    prefix = 'expanded_conv_{}_'.format(block_id)
    if block_id:
        # Expand

        x = Conv2D(expansion * in_channels, kernel_size=1, padding='same',
                   use_bias=False, activation=None,
                   name=prefix + 'expand')(x)
        x = BatchNormalization(epsilon=1e-3, momentum=0.999,
                               name=prefix + 'expand_BN')(x)
        x = Activation(relu6, name=prefix + 'expand_relu')(x)
    else:
        prefix = 'expanded_conv_'
    # Depthwise
    x = DepthwiseConv2D(kernel_size=3, strides=stride, activation=None,
                        use_bias=False, padding='same', dilation_rate=(rate, rate),
                        name=prefix + 'depthwise')(x)
    x = BatchNormalization(epsilon=1e-3, momentum=0.999,
                           name=prefix + 'depthwise_BN')(x)

    x = Activation(relu6, name=prefix + 'depthwise_relu')(x)

    # Project
    x = Conv2D(pointwise_filters,
               kernel_size=1, padding='same', use_bias=False, activation=None,
               name=prefix + 'project')(x)
    x = BatchNormalization(epsilon=1e-3, momentum=0.999,
                           name=prefix + 'project_BN')(x)

    if skip_connection:
        return Add(name=prefix + 'add')([inputs, x])

    # if in_channels == pointwise_filters and stride == 1:
    #    return Add(name='res_connect_' + str(block_id))([inputs, x])

    return x 

def _xception_block(inputs, depth_list, prefix, skip_connection_type, stride,
                    rate=1, depth_activation=False, return_skip=False):
    """ Basic building block of modified Xception network
        Args:
            inputs: input tensor
            depth_list: number of filters in each SepConv layer. len(depth_list) == 3
            prefix: prefix before name
            skip_connection_type: one of {'conv','sum','none'}
            stride: stride at last depthwise conv
            rate: atrous rate for depthwise convolution
            depth_activation: flag to use activation between depthwise & pointwise convs
            return_skip: flag to return additional tensor after 2 SepConvs for decoder
            """
    residual = inputs
    for i in range(3):
        residual = SepConv_BN(residual,
                              depth_list[i],
                              prefix + '_separable_conv{}'.format(i + 1),
                              stride=stride if i == 2 else 1,
                              rate=rate,
                              depth_activation=depth_activation)
        if i == 1:
            skip = residual
    if skip_connection_type == 'conv':
        shortcut = _conv2d_same(inputs, depth_list[-1], prefix + '_shortcut',
                                kernel_size=1,
                                stride=stride)
        shortcut = BatchNormalization(name=prefix + '_shortcut_BN')(shortcut)
        outputs = layers.add([residual, shortcut])
    elif skip_connection_type == 'sum':
        outputs = layers.add([residual, inputs])
    elif skip_connection_type == 'none':
        outputs = residual
    if return_skip:
        return outputs, skip
    else:
        return outputs 

def SepConv_BN(x, filters, prefix, stride=1, kernel_size=3, rate=1, depth_activation=False, epsilon=1e-3):
    """ SepConv with BN between depthwise & pointwise. Optionally add activation after BN
        Implements right "same" padding for even kernel sizes
        Args:
            x: input tensor
            filters: num of filters in pointwise convolution
            prefix: prefix before name
            stride: stride at depthwise conv
            kernel_size: kernel size for depthwise convolution
            rate: atrous rate for depthwise convolution
            depth_activation: flag to use activation between depthwise & poinwise convs
            epsilon: epsilon to use in BN layer
    """

    if stride == 1:
        depth_padding = 'same'
    else:
        kernel_size_effective = kernel_size + (kernel_size - 1) * (rate - 1)
        pad_total = kernel_size_effective - 1
        pad_beg = pad_total // 2
        pad_end = pad_total - pad_beg
        x = ZeroPadding2D((pad_beg, pad_end))(x)
        depth_padding = 'valid'

    if not depth_activation:
        x = Activation('relu')(x)
    x = DepthwiseConv2D((kernel_size, kernel_size), strides=(stride, stride), dilation_rate=(rate, rate),
                        padding=depth_padding, use_bias=False, name=prefix + '_depthwise')(x)
    x = BatchNormalization(name=prefix + '_depthwise_BN', epsilon=epsilon)(x)
    if depth_activation:
        x = Activation('relu')(x)
    x = Conv2D(filters, (1, 1), padding='same',
               use_bias=False, name=prefix + '_pointwise')(x)
    x = BatchNormalization(name=prefix + '_pointwise_BN', epsilon=epsilon)(x)
    if depth_activation:
        x = Activation('relu')(x)

    return x 

def ResPath(filters, length, inp):
    '''
    ResPath

    Arguments:
        filters {int} -- [description]
        length {int} -- length of ResPath
        inp {keras layer} -- input layer

    Returns:
        [keras layer] -- [output layer]
    '''

    shortcut = inp
    shortcut = conv3d_bn(shortcut, filters , 1, 1, 1, activation=None, padding='same')

    out = conv3d_bn(inp, filters, 3, 3, 3, activation='relu', padding='same')

    out = add([shortcut, out])
    out = Activation('relu')(out)
    out = BatchNormalization(axis=4)(out)

    for i in range(length-1):

        shortcut = out
        shortcut = conv3d_bn(shortcut, filters , 1, 1, 1, activation=None, padding='same')

        out = conv3d_bn(out, filters, 3, 3, 3, activation='relu', padding='same')

        out = add([shortcut, out])
        out = Activation('relu')(out)
        out = BatchNormalization(axis=4)(out)


    return out 

def MultiResBlock(U, inp, alpha = 1.67):
    '''
    MultiRes Block

    Arguments:
        U {int} -- Number of filters in a corrsponding UNet stage
        inp {keras layer} -- input layer

    Returns:
        [keras layer] -- [output layer]
    '''

    W = alpha * U

    shortcut = inp

    shortcut = conv3d_bn(shortcut, int(W*0.167) + int(W*0.333) + int(W*0.5), 1, 1, 1, activation=None, padding='same')

    conv3x3 = conv3d_bn(inp, int(W*0.167), 3, 3, 3, activation='relu', padding='same')

    conv5x5 = conv3d_bn(conv3x3, int(W*0.333), 3, 3, 3, activation='relu', padding='same')

    conv7x7 = conv3d_bn(conv5x5, int(W*0.5), 3, 3, 3, activation='relu', padding='same')

    out = concatenate([conv3x3, conv5x5, conv7x7], axis=4)
    out = BatchNormalization(axis=4)(out)

    out = add([shortcut, out])
    out = Activation('relu')(out)
    out = BatchNormalization(axis=4)(out)

    return out 

def apn_module(self, x):

        def right(x):
            x = layers.AveragePooling2D()(x)
            x = layers.Conv2D(self.classes, kernel_size=1, padding='same')(x)
            x = layers.BatchNormalization()(x)
            x = layers.Activation('relu')(x)
            x = layers.UpSampling2D(interpolation='bilinear')(x)
            return x

        def conv(x, filters, kernel_size, stride):
            x = layers.Conv2D(filters, kernel_size=kernel_size, strides=(stride, stride), padding='same')(x)
            x = layers.BatchNormalization()(x)
            x = layers.Activation('relu')(x)
            return x

        x_7 = conv(x, int(x.shape[-1]), 7, stride=2)
        x_5 = conv(x_7, int(x.shape[-1]), 5, stride=2)
        x_3 = conv(x_5, int(x.shape[-1]), 3, stride=2)

        x_3_1 = conv(x_3, self.classes, 3, stride=1)
        x_3_1_up = layers.UpSampling2D(interpolation='bilinear')(x_3_1)
        x_5_1 = conv(x_5, self.classes, 5, stride=1)
        x_3_5 = layers.add([x_5_1, x_3_1_up])
        x_3_5_up = layers.UpSampling2D(interpolation='bilinear')(x_3_5)
        x_7_1 = conv(x_7, self.classes, 3, stride=1)
        x_3_5_7 = layers.add([x_7_1, x_3_5_up])
        x_3_5_7_up = layers.UpSampling2D(interpolation='bilinear')(x_3_5_7)

        x_middle = conv(x, self.classes, 1, stride=1)
        x_middle = layers.multiply([x_3_5_7_up, x_middle])

        x_right = right(x)
        x_middle = layers.add([x_middle, x_right])
        return x_middle