我真服了原来我之前用tf复现SegNet给复现错了
在网上试了多个版本代码，折腾了好久，现在终于复现对了，代码也跑通了
SegNet的架构比较老了，这几年都没人更新代码了，我这里算是提供一个最近能跑通的版本的代码吧
tf版本2.4.1
首先主要是构建两个类来实现池化索引，这里经过反复尝试我懵懵懂懂地解决了其它代码直接搬运过来导致的各种报错
import tensorflow as tf
from tensorflow.keras import backend as K
from tensorflow.keras.layers import Layerclass MaxPoolingWithArgmax2D(Layer):def __init__(self, pool_size=(2, 2), strides=(2, 2), padding='same', **kwargs):super(MaxPoolingWithArgmax2D, self).__init__(**kwargs)self.padding = paddingself.pool_size = pool_sizeself.strides = stridesdef call(self, inputs, **kwargs):padding = self.paddingpool_size = self.pool_sizestrides = self.stridesif K.backend() == 'tensorflow':ksize = [1, pool_size[0], pool_size[1], 1]padding = padding.upper()strides = [1, strides[0], strides[1], 1]output, argmax = tf.nn.max_pool_with_argmax(inputs, ksize=ksize, strides=strides, padding=padding)else:errmsg = '{} backend is not supported for layer {}'.format(K.backend(), type(self).__name__)raise NotImplementedError(errmsg)argmax = K.cast(argmax, K.floatx())return [output, argmax]def compute_output_shape(self, input_shape):ratio = (1, 2, 2, 1)output_shape = [dim // ratio[idx] if dim is not None else None for idx, dim in enumerate(input_shape)]output_shape = tuple(output_shape)return [output_shape, output_shape]def compute_mask(self, inputs, mask=None):return 2 * [None]def get_config(self):config = super(MaxPoolingWithArgmax2D, self).get_config()config.update({"pool_size": self.pool_size,"strides": self.strides,"padding": self.padding,})return configclass MaxUnpooling2D(Layer):def __init__(self, size=(2, 2), **kwargs):super(MaxUnpooling2D, self).__init__(**kwargs)self.size = sizedef call(self, inputs, output_shape=None):updates, mask = inputs[0], inputs[1]with tf.compat.v1.variable_scope(self.name):mask = K.cast(mask, 'int32')input_shape = tf.shape(updates, out_type='int32')# calculation new shapeif output_shape is None:output_shape = (input_shape[0], input_shape[1] * self.size[0], input_shape[2] * self.size[1], input_shape[3])self.output_shape1 = output_shape# calculation indices for batch, height, width and feature mapsone_like_mask = K.ones_like(mask, dtype='int32')batch_shape = K.concatenate([[input_shape[0]], [1], [1], [1]], axis=0)batch_range = K.reshape(tf.range(output_shape[0], dtype='int32'), shape=batch_shape)b = one_like_mask * batch_rangey = mask // (output_shape[2] * output_shape[3])x = (mask // output_shape[3]) % output_shape[2]feature_range = tf.range(output_shape[3], dtype='int32')f = one_like_mask * feature_range# transpose indices & reshape update values to one dimensionupdates_size = tf.size(updates)indices = K.transpose(K.reshape(K.stack([b, y, x, f]), [4, updates_size]))values = K.reshape(updates, [updates_size])ret = tf.scatter_nd(indices, values, output_shape)input_shape = updates.shapeout_shape = [-1,input_shape[1] * self.size[0],input_shape[2] * self.size[1],input_shape[3]]return K.reshape(ret, out_shape)def compute_output_shape(self, input_shape):mask_shape = input_shape[1]return mask_shape[0], mask_shape[1] * self.size[0], mask_shape[2] * self.size[1], mask_shape[3]def get_config(self):config = super(MaxUnpooling2D, self).get_config()config.update({"size": self.size,})return config
另外SegNet网络主体部分，注意池化和反池化的时候filters数量要对得上
def SegNet(fNum, dates, lossweights, filters=64):inputs = keras.layers.Input((fNum*dates, img_h, img_w))inputs0 = keras.layers.Lambda(reshapes2)(inputs) # 针对我数据的reshape# Encoderconv1 = keras.layers.Conv2D(filters, (3, 3), activation='relu', padding='same')(inputs0)conv1 = keras.layers.BatchNormalization()(conv1)conv1 = keras.layers.Conv2D(filters, (3, 3), activation='relu', padding='same')(conv1)conv1 = keras.layers.BatchNormalization()(conv1)pool1, idx1 = MaxPoolingWithArgmax2D(pool_size=(2, 2))(conv1)conv2 = keras.layers.Conv2D(filters*2, (3, 3), activation='relu', padding='same')(pool1)conv2 = keras.layers.BatchNormalization()(conv2)conv2 = keras.layers.Conv2D(filters*2, (3, 3), activation='relu', padding='same')(conv2)conv2 = keras.layers.BatchNormalization()(conv2)pool2, idx2 = MaxPoolingWithArgmax2D(pool_size=(2, 2))(conv2)conv3 = keras.layers.Conv2D(filters*4, (3, 3), activation='relu', padding='same')(pool2)conv3 = keras.layers.BatchNormalization()(conv3)conv3 = keras.layers.Conv2D(filters*4, (3, 3), activation='relu', padding='same')(conv3)conv3 = keras.layers.BatchNormalization()(conv3)pool3, idx3 = MaxPoolingWithArgmax2D(pool_size=(2, 2))(conv3)conv4 = keras.layers.Conv2D(filters*8, (3, 3), activation='relu', padding='same')(pool3)conv4 = keras.layers.BatchNormalization()(conv4)conv4 = keras.layers.Conv2D(filters*8, (3, 3), activation='relu', padding='same')(conv4)conv4 = keras.layers.BatchNormalization()(conv4)pool4, idx4 = MaxPoolingWithArgmax2D(pool_size=(2, 2))(conv4)# Decoderup5 = MaxUnpooling2D((2,2))([pool4, idx4])conv5 = keras.layers.Conv2D(filters*4, (3, 3), activation='relu', padding='same')(up5)conv5 = keras.layers.BatchNormalization()(conv5)conv5 = keras.layers.Conv2D(filters*4, (3, 3), activation='relu', padding='same')(conv5)conv5 = keras.layers.BatchNormalization()(conv5)up6 = MaxUnpooling2D(size=(2, 2))([conv5, idx3])conv6 = keras.layers.Conv2D(filters*2, (3, 3), activation='relu', padding='same')(up6)conv6 = keras.layers.BatchNormalization()(conv6)conv6 = keras.layers.Conv2D(filters*2, (3, 3), activation='relu', padding='same')(conv6)conv6 = keras.layers.BatchNormalization()(conv6)up7 = MaxUnpooling2D(size=(2, 2))([conv6, idx2])conv7 = keras.layers.Conv2D(filters, (3, 3), activation='relu', padding='same')(up7)conv7 = keras.layers.BatchNormalization()(conv7)conv7 = keras.layers.Conv2D(filters, (3, 3), activation='relu', padding='same')(conv7)conv7 = keras.layers.BatchNormalization()(conv7)up8 = MaxUnpooling2D(size=(2, 2))([conv7, idx1])conv8 = keras.layers.Conv2D(16, (3, 3), activation='relu', padding='same')(up8)conv8 = keras.layers.BatchNormalization()(conv8)conv8 = keras.layers.Conv2D(16, (3, 3), activation='relu', padding='same')(conv8)conv8 = keras.layers.BatchNormalization()(conv8)outputs = keras.layers.Conv2D(1, (1, 1), activation='sigmoid')(conv8)model = keras.models.Model(inputs=inputs, outputs=outputs)return model
