nasnet.py 文件源码

def _reduction_A(ip, p, filters, weight_decay=5e-5, id=None):
    '''Adds a Reduction cell for NASNet-A (Fig. 4 in the paper)

    # Arguments:
        ip: input tensor `x`
        p: input tensor `p`
        filters: number of output filters
        weight_decay: l2 regularization weight
        id: string id

    # Returns:
        a Keras tensor
    '''
    """"""
    channel_dim = 1 if K.image_data_format() == 'channels_first' else -1

    with K.name_scope('reduction_A_block_%s' % id):
        p = _adjust_block(p, ip, filters, weight_decay, id)

        h = Activation('relu')(ip)
        h = Conv2D(filters, (1, 1), strides=(1, 1), padding='same', name='reduction_conv_1_%s' % id,
                   use_bias=False, kernel_initializer='he_normal', kernel_regularizer=l2(weight_decay))(h)
        h = BatchNormalization(axis=channel_dim, momentum=_BN_DECAY, epsilon=_BN_EPSILON,
                               name='reduction_bn_1_%s' % id)(h)

        with K.name_scope('block_1'):
            x1_1 = _separable_conv_block(h, filters, (5, 5), strides=(2, 2), weight_decay=weight_decay,
                                         id='reduction_left1_%s' % id)
            x1_2 = _separable_conv_block(p, filters, (7, 7), strides=(2, 2), weight_decay=weight_decay,
                                         id='reduction_1_%s' % id)
            x1 = add([x1_1, x1_2], name='reduction_add_1_%s' % id)

        with K.name_scope('block_2'):
            x2_1 = MaxPooling2D((3, 3), strides=(2, 2), padding='same', name='reduction_left2_%s' % id)(h)
            x2_2 = _separable_conv_block(p, filters, (7, 7), strides=(2, 2), weight_decay=weight_decay,
                                         id='reduction_right2_%s' % id)
            x2 = add([x2_1, x2_2], name='reduction_add_2_%s' % id)

        with K.name_scope('block_3'):
            x3_1 = AveragePooling2D((3, 3), strides=(2, 2), padding='same', name='reduction_left3_%s' % id)(h)
            x3_2 = _separable_conv_block(p, filters, (5, 5), strides=(2, 2), weight_decay=weight_decay,
                                         id='reduction_right3_%s' % id)
            x3 = add([x3_1, x3_2], name='reduction_add3_%s' % id)

        with K.name_scope('block_4'):
            x4 = AveragePooling2D((3, 3), strides=(1, 1), padding='same', name='reduction_left4_%s' % id)(x1)
            x4 = add([x2, x4])

        with K.name_scope('block_5'):
            x5_1 = _separable_conv_block(x1, filters, (3, 3), weight_decay=weight_decay, id='reduction_left4_%s' % id)
            x5_2 = MaxPooling2D((3, 3), strides=(2, 2), padding='same', name='reduction_right5_%s' % id)(h)
            x5 = add([x5_1, x5_2], name='reduction_add4_%s' % id)

        x = concatenate([x2, x3, x4, x5], axis=channel_dim, name='reduction_concat_%s' % id)
        return x, ip