def model_simple_upsampling__reshape(img_shape, class_n=None):
from keras.layers import Input, Dense, Convolution3D, MaxPooling3D, UpSampling3D, Reshape, Flatten
from keras.models import Sequential, Model
from keras.layers.core import Activation
from aitom.classify.deep.unsupervised.autoencoder.seg_util import conv_block
NUM_CHANNELS = 1
input_shape = (None, img_shape[0], img_shape[1], img_shape[2],
NUM_CHANNELS)
# use relu activation for hidden layer to guarantee non-negative outputs are passed
# to the max pooling layer. In such case, as long as the output layer is linear activation,
# the network can still accomodate negative image intendities,
# just matter of shift back using the bias term
input_img = Input(shape=input_shape[1:])
x = input_img
x = conv_block(x, 32, 3, 3, 3)
x = MaxPooling3D((2, 2, 2), border_mode='same')(x)
x = conv_block(x, 32, 3, 3, 3)
x = MaxPooling3D((2, 2, 2), border_mode='same')(x)
x = conv_block(x, 32, 3, 3, 3)
x = UpSampling3D((2, 2, 2))(x)
x = conv_block(x, 32, 3, 3, 3)
x = UpSampling3D((2, 2, 2))(x)
x = conv_block(x, 32, 3, 3, 3)
x = Convolution3D(class_n, 1, 1, 1, border_mode='same')(x)
x = Reshape((N.prod(img_shape), class_n))(x)
x = Activation('softmax')(x)
model = Model(input=input_img, output=x)
print('model layers:')
for l in model.layers:
print(l.output_shape, l.name)
return model
def encoder_simple_conv(img_shape, encoding_dim=32, NUM_CHANNELS=1):
# workaround for Dropout to work
#import tensorflow as tf
#tf.python.control_flow_ops = tf
# from seg_util import conv_block
from aitom.classify.deep.unsupervised.autoencoder.seg_util import conv_block
from keras.layers import Input, Dense, Convolution3D, MaxPooling3D, UpSampling3D, Reshape, Flatten
from keras.models import Sequential, Model
from keras import regularizers
input_shape = (None, img_shape[0], img_shape[1], img_shape[2], NUM_CHANNELS)
# use relu activation for hidden layer to guarantee non-negative outputs are passed to the max pooling layer. In such case, as long as the output layer is linear activation, the network can still accomodate negative image intendities, just matter of shift back using the bias term
input_img = Input(shape=input_shape[1:])
x = input_img
x = conv_block(x, 32, 3, 3, 3)
x = MaxPooling3D((2, 2, 2), border_mode='same')(x)
x = conv_block(x, 32, 3, 3, 3)
x = MaxPooling3D((2, 2, 2), border_mode='same')(x)
encoder_conv_shape = [_.value for _ in x.get_shape()] # x.get_shape() returns a list of tensorflow.python.framework.tensor_shape.Dimension objects
x = Flatten()(x)
if False:
x = Dense(encoding_dim, activation='relu')(x)
else:
x = Dense(encoding_dim, activation='relu', activity_regularizer=regularizers.l1(10e-5))(x) # with sparsity
encoded = x
encoder = Model(input=input_img, output=encoded)
print ('encoder', 'input shape', encoder.input_shape, 'output shape', encoder.output_shape)
input_img_decoder = Input(shape=encoder.output_shape[1:])
x = input_img_decoder
x = Dense(N.prod(encoder_conv_shape[1:]), activation='relu')(x)
x = Reshape(encoder_conv_shape[1:])(x)
x = UpSampling3D((2, 2, 2))(x)
x = conv_block(x, 32, 3, 3, 3)
x = UpSampling3D((2, 2, 2))(x)
x = conv_block(x, 32, 3, 3, 3)
x = Convolution3D(1, 3, 3, 3, activation='linear', border_mode='same')(x) # keep the output layer linear activation, so that the image intensity can be negative
decoded = x
decoder = Model(input=input_img_decoder, output=decoded)
autoencoder = Sequential()
if True:
# model with expanded layers, and hopefully allow parallel training
for l in encoder.layers: autoencoder.add(l)
for l in decoder.layers: autoencoder.add(l)
else:
# build encoder according to ~/src/tmp/proj/gan/dcgan.py
autoencoder.add(encoder)
autoencoder.add(decoder)
print('autoencoder layers:')
for l in autoencoder.layers: print (l.output_shape)
return {'autoencoder':autoencoder, 'encoder':encoder, 'decoder':decoder}