python - 将 2DResnet 覆盖到 3DResnet
问题描述
我想将 2D resnet 转换为 3D resnet 并遵循此处的结构/格式,因为这似乎是我能找到的最简单的示例: https ://github.com/priya-dwivedi/Deep-Learning/blob/master/ resnet_keras/Residual_Networks_yourself.ipynb
我将 conv2D 更改为 conv3D,MaxPooling2D 更改为 MaxPooling3D 等,并将维度更新为 3D 输入数组。我还减少了过滤器的数量,因为我没有大型数据集,并且希望尽可能保持简单。
但是模型摘要对我来说看起来不正确,它似乎从正确的图像尺寸 84x84x84x1 开始,但是对于 resnet,过滤器尺寸似乎不正确?例如,此链接中显示的模型摘要更具可读性和清晰性:https ://github.com/ageron/handson-ml2/issues/48
任何对 ResNets 有更多经验的人都可以提供一些指导吗?这是您将 2D 转换为 3D resnet 的方式吗?基于以下实现,模型摘要对于 84x84x84x1 的输入图像是否正确?
我的 2D 到 3D 代码:
def identity_block(X, f, filters, stage, block):
"""
Implementation of the identity block as defined in Figure 3
Arguments:
X -- input tensor of shape (m, n_H_prev, n_W_prev, n_C_prev)
f -- integer, specifying the shape of the middle CONV's window for the main path
filters -- python list of integers, defining the number of filters in the CONV layers of the main path
stage -- integer, used to name the layers, depending on their position in the network
block -- string/character, used to name the layers, depending on their position in the network
Returns:
X -- output of the identity block, tensor of shape (n_H, n_W, n_C)
"""
# defining name basis
conv_name_base = 'res' + str(stage) + block + '_branch'
bn_name_base = 'bn' + str(stage) + block + '_branch'
# Retrieve Filters
F1, F2, F3 = filters
# Save the input value. You'll need this later to add back to the main path.
X_shortcut = X
# First component of main path
X = Conv3D(filters = F1, kernel_size = (1, 1, 1), strides = (1,1, 1), padding = 'valid', name = conv_name_base + '2a', kernel_initializer = glorot_uniform(seed=0))(X)
X = BatchNormalization(name = bn_name_base + '2a')(X)
X = Activation('relu')(X)
# Second component of main path (≈3 lines)
X = Conv3D(filters = F2, kernel_size = (f, f, f), strides = (1,1,1), padding = 'same', name = conv_name_base + '2b', kernel_initializer = glorot_uniform(seed=0))(X)
X = BatchNormalization(name = bn_name_base + '2b')(X)
X = Activation('relu')(X)
# Third component of main path (≈2 lines)
X = Conv3D(filters = F3, kernel_size = (1, 1, 1), strides = (1,1, 1), padding = 'valid', name = conv_name_base + '2c', kernel_initializer = glorot_uniform(seed=0))(X)
X = BatchNormalization(name = bn_name_base + '2c')(X)
# Final step: Add shortcut value to main path, and pass it through a RELU activation (≈2 lines)
X = Add()([X, X_shortcut])
X = Activation('relu')(X)
return X
def convolutional_block(X, f, filters, stage, block, s = 2):
"""
Arguments:
X -- input tensor of shape (m, n_H_prev, n_W_prev, n_C_prev)
f -- integer, specifying the shape of the middle CONV's window for the main path
filters -- python list of integers, defining the number of filters in the CONV layers of the main path
stage -- integer, used to name the layers, depending on their position in the network
block -- string/character, used to name the layers, depending on their position in the network
s -- Integer, specifying the stride to be used
Returns:
X -- output of the convolutional block, tensor of shape (n_H, n_W, n_C)
"""
# defining name basis
conv_name_base = 'res' + str(stage) + block + '_branch'
bn_name_base = 'bn' + str(stage) + block + '_branch'
# Retrieve Filters
F1, F2, F3 = filters
# Save the input value
X_shortcut = X
##### MAIN PATH #####
# First component of main path
X = Conv3D(F1, (1, 1, 1), strides = (s,s,s), name = conv_name_base + '2a', kernel_initializer = glorot_uniform(seed=0))(X)
X = BatchNormalization(name = bn_name_base + '2a')(X)
X = Activation('relu')(X)
# Second component of main path (≈3 lines)
X = Conv3D(filters = F2, kernel_size = (f, f, f), strides = (1, 1, 1), padding = 'same', name = conv_name_base + '2b', kernel_initializer = glorot_uniform(seed=0))(X)
X = BatchNormalization(name = bn_name_base + '2b')(X)
X = Activation('relu')(X)
# Third component of main path (≈2 lines)
X = Conv3D(filters = F3, kernel_size = (1, 1, 1), strides = (1,1, 1), padding = 'valid', name = conv_name_base + '2c', kernel_initializer = glorot_uniform(seed=0))(X)
X = BatchNormalization(name = bn_name_base + '2c')(X)
##### SHORTCUT PATH #### (≈2 lines)
X_shortcut = Conv3D(filters = F3, kernel_size = (1, 1, 1), strides = (s,s,s), padding = 'valid', name = conv_name_base + '1',
kernel_initializer = glorot_uniform(seed=0))(X_shortcut)
X_shortcut = BatchNormalization(name = bn_name_base + '1')(X_shortcut)
# Final step: Add shortcut value to main path, and pass it through a RELU activation (≈2 lines)
X = Add()([X, X_shortcut])
X = Activation('relu')(X)
return X
def ResNet50(input_shape=(82, 82,82, 1), classes=3):
"""
Implementation of the popular ResNet50 the following architecture:
CONV2D -> BATCHNORM -> RELU -> MAXPOOL -> CONVBLOCK -> IDBLOCK*2 -> CONVBLOCK -> IDBLOCK*3
-> CONVBLOCK -> IDBLOCK*5 -> CONVBLOCK -> IDBLOCK*2 -> AVGPOOL -> TOPLAYER
Arguments:
input_shape -- shape of the images of the dataset
classes -- integer, number of classes
Returns:
model -- a Model() instance in Keras
"""
# Define the input as a tensor with shape input_shape
X_input = Input(input_shape)
# Zero-Padding
X = ZeroPadding3D((3, 3, 3))(X_input)
# Stage 1
X = Conv3D(32, (7, 7, 7), strides=(2, 2, 2), name='conv1', kernel_initializer=glorot_uniform(seed=0))(X)
X = BatchNormalization(name='bn_conv1')(X)
X = Activation('relu')(X)
X = MaxPooling3D((3, 3, 3), strides=(2, 2, 2))(X)
# Stage 2
X = convolutional_block(X, f=3, filters=[32, 32, 64], stage=2, block='a', s=1)
X = identity_block(X, 3, [32, 32, 64], stage=2, block='b')
X = identity_block(X, 3, [32, 32, 64], stage=2, block='c')
# Stage 3 (≈4 lines)
X = convolutional_block(X, f = 3, filters = [64, 64, 128], stage = 3, block='a', s = 2)
X = identity_block(X, 3, [64, 64, 128], stage=3, block='b')
X = identity_block(X, 3, [64, 64, 128], stage=3, block='c')
X = identity_block(X, 3, [64, 64, 128], stage=3, block='d')
# Stage 4 (≈6 lines)
X = convolutional_block(X, f = 3, filters = [128, 128, 256], stage = 4, block='a', s = 2)
X = identity_block(X, 3, [128, 128, 256], stage=4, block='b')
X = identity_block(X, 3, [128, 128, 256], stage=4, block='c')
X = identity_block(X, 3, [128, 128, 256], stage=4, block='d')
X = identity_block(X, 3, [128, 128, 256], stage=4, block='e')
X = identity_block(X, 3, [128, 128, 256], stage=4, block='f')
# Stage 5 (≈3 lines)
X = convolutional_block(X, f = 3, filters = [256, 256, 512], stage = 5, block='a', s = 2)
X = identity_block(X, 3, [256, 256, 512], stage=5, block='b')
X = identity_block(X, 3, [256, 256, 512], stage=5, block='c')
# AVGPOOL (≈1 line). Use "X = AveragePooling2D(...)(X)"
X = AveragePooling3D((2,2,2), name="avg_pool")(X)
### END CODE HERE ###
# output layer
X = Flatten()(X)
X = Dense(classes, activation='softmax', name='fc' + str(classes), kernel_initializer = glorot_uniform(seed=0))(X)
# Create model
model = Model(inputs = X_input, outputs = X, name='ResNet50')
return model
model = ResNet50(input_shape = (82, 82,82, 1), classes = 3)
model.summary()
解决方案
推荐阅读
- java - POI中如何设置单元格数据类型?
- react-native - 如何关闭选项卡上的相机或查看更改
- nativescript - 如何在模态中显示原始图像大小?
- kubernetes - 如何检查我的 Kubernetes 集群是否有资源来部署我的所有软件
- python - 如何根据 URL 列表制作嵌套字典?
- apache-pulsar - Apache Pulsar - ioThreads/listenerThreads 和消息排序
- normalization - 如何在 Jbuilder 和 Normalizr 之间展平数据
- c# - WebSocketSharp 异常:System.PlatformNotSupportedException:“此平台不支持操作。”
- firefox - Nightwatch FireFox60 和 geckodriver MoveTo 因未知命令而失败
- html - 为什么行从 div VERTICALLY 溢出?