def UpsampleConv(name, input_dim, output_dim, filter_size, inputs, he_init=True, biases=True): output = inputs output = tf.concat([output, output, output, output], axis=1) output = tf.transpose(output, [0,2,3,1]) output = tf.depth_to_space(output, 2) output = tf.transpose(output, [0,3,1,2]) output = Conv2D(name, input_dim, output_dim, filter_size, output, he_init=he_init, biases=biases, cpu=CPU) return output
def resnet_decoder(name, z, n_channels, is_training, mode=None, nonlinearity=tf.nn.relu): #print('In decoder network') output = Linear(name + '.Input', z.get_shape().as_list()[1], 4 * 4 * 8 * DIM, z) #print output.shape output = Normalize(name + '.BN0', [0], output, is_training, mode) #print output.shape output = nonlinearity(output) #print output.shape output = tf.reshape(output, [-1, 8 * DIM, 4, 4]) #print output.shape #output = ResidualBlock(name + '.Res1', 8*DIM, 8*DIM, 3, output, is_training, mode, resample='up', norm_inputs=True) #output = ResidualBlock(name + '.Res2', 8*DIM, 4*DIM, 3, output, is_training, mode, resample='up') output = ResidualBlock(name + '.Res1', 8 * DIM, 4 * DIM, 3, output, is_training, mode, resample='up', norm_inputs=True) #print output.shape output = ResidualBlock(name + '.Res2', 4 * DIM, 2 * DIM, 3, output, is_training, mode, resample='up') #print output.shape output = ResidualBlock(name + '.Res3', 2 * DIM, 1 * DIM, 3, output, is_training, mode, resample='up') #print output.shape output = Normalize(name + '.BN5', [0, 2, 3], output, is_training, mode) #print output.shape output = nonlinearity(output) #print output.shape output = Conv2D(name + '.Output', DIM, n_channels, 3, output) #print output.shape output = tf.reshape(output, [-1, n_channels * DIM * DIM]) #print output.shape return output
def resnet_decoder(name, z, n_channels, is_training, mode=None, nonlinearity=tf.nn.relu): output = Linear(name + '.Input', z.get_shape().as_list()[1], (DIM / 16) * (DIM / 16) * 8 * DIM, z) output = Normalize(name + '.BN0', [0], output, is_training, mode) output = nonlinearity(output) output = tf.reshape(output, [-1, 8 * DIM, (DIM / 16), (DIM / 16)]) output = ResidualBlock(name + '.Res1', 8 * DIM, 8 * DIM, 3, output, is_training, mode, resample='up', norm_inputs=True) output = ResidualBlock(name + '.Res2', 8 * DIM, 4 * DIM, 3, output, is_training, mode, resample='up') output = ResidualBlock(name + '.Res3', 4 * DIM, 2 * DIM, 3, output, is_training, mode, resample='up') output = ResidualBlock(name + '.Res4', 2 * DIM, 1 * DIM, 3, output, is_training, mode, resample='up') output = Normalize(name + '.BN5', [0, 2, 3], output, is_training, mode) output = nonlinearity(output) output = Conv2D(name + '.Output', DIM, n_channels, 3, output) output = tf.reshape(output, [-1, n_channels * DIM * DIM]) return output
def resnet_encoder_new(name, inputs, n_channels, latent_dim, is_training, mode=None, nonlinearity=tf.nn.relu): output = tf.reshape(inputs, [-1, n_channels, DIM, DIM]) output = Conv2D(name + '.Input', n_channels, DIM, 3, output, he_init=False, cpu=CPU) output = ResidualBlock(name + '.Res1', DIM, 2 * DIM, 3, output, is_training, mode, resample='down') output = ResidualBlock(name + '.Res2', 2 * DIM, 4 * DIM, 3, output, is_training, mode, resample='down') output = ResidualBlock(name + '.Res3', 4 * DIM, 8 * DIM, 3, output, is_training, mode, resample='down') output = ResidualBlock(name + '.Res4', 8 * DIM, 8 * DIM, 3, output, is_training, mode, resample='down') output = Normalize(name + '.BN.5', [0, 2, 3], output, is_training, mode) output = nonlinearity(output) output = tf.reshape(output, [-1, 4 * 4 * 8 * DIM]) output = Linear(name + '.Output', 4 * 4 * 8 * DIM, latent_dim, output) return output
def resnet_encoder(name, inputs, n_channels, latent_dim, is_training, mode=None, nonlinearity=tf.nn.relu): output = tf.reshape(inputs, [-1, n_channels, DIM, DIM]) #Conv2D(name, input_dim, output_dim, filter_size, inputs, he_init=True, mask_type=None, stride=1, weightnorm=None, biases=True, gain=1., cpu=False): output = Conv2D(name + '.Input', n_channels, DIM, 3, output, he_init=False, cpu=CPU) output = ResidualBlock(name + '.Res1', DIM, 2 * DIM, 3, output, is_training, mode, resample='down') output = ResidualBlock(name + '.Res2', 2 * DIM, 4 * DIM, 3, output, is_training, mode, resample='down') output = ResidualBlock(name + '.Res3', 4 * DIM, 8 * DIM, 3, output, is_training, mode, resample='down') output = ResidualBlock(name + '.Res4', 8 * DIM, 8 * DIM, 3, output, is_training, mode, resample='down') output = tf.reshape(output, [-1, (DIM / 16) * (DIM / 16) * 8 * DIM]) # output = Linear(name + '.Output', (DIM / 16) * (DIM / 16) * 8 * DIM, latent_dim, output) return output
def high_capacity_encoder(name, inputs, n_channels, latent_dim, is_training, mode=None, nonlinearity=tf.nn.relu): output = tf.reshape(inputs, [-1, n_channels, DIM, DIM]) output = Conv2D(name + '.0', n_channels, DIM, 5, output, stride=2) output = Normalize(name + '.BN0', [0,2,3], output, is_training, mode) output = nonlinearity(output) output = Conv2D(name + '.1', DIM, DIM*2, 5, output, stride=2) output = Normalize(name + '.BN1', [0,2,3], output, is_training, mode) output = nonlinearity(output) output = Conv2D(name + '.2', DIM*2, DIM*4, 5, output, stride=2) output = Normalize(name + '.BN2', [0,2,3], output, is_training, mode) output = nonlinearity(output) output = tf.reshape(output, [-1, DIM*4*8*8]) output = Linear(name + '.FC', DIM*4*8*8, DIM*4*8, output) output = Normalize(name + '.BNFC', [0], output, is_training, mode) output = nonlinearity(output) output = Linear(name + '.Output', DIM*4*8, latent_dim, output) return output
def gan_discriminator(name, inputs, n_channels, is_training, mode=None, nonlinearity=LeakyReLU): output = tf.reshape(inputs, [-1, n_channels, DIM, DIM]) output = Conv2D(name + '.discriminator1', n_channels, DIM, 5, output, stride=2) output = nonlinearity(output) output = Conv2D(name + '.discriminator2', DIM, 2 * DIM, 5, output, stride=2) output = nonlinearity(output) output = Conv2D(name + '.discriminator3', 2 * DIM, 4 * DIM, 5, output, stride=2) output = nonlinearity(output) output = Conv2D(name + '.discriminator4', 4 * DIM, 4 * DIM, 5, output, stride=2) output = nonlinearity(output) output = tf.reshape(output, [-1, 4 * 4 * 4 * DIM]) output = Linear(name + '.Output', 4 * 4 * 4 * DIM, 1, output) return output
def resnet_encoder(name, inputs, n_channels, latent_dim, is_training, mode=None, nonlinearity=tf.nn.relu): #print('in encoder network') #print inputs.shape output = tf.reshape(inputs, [-1, n_channels, DIM, DIM]) #print output.shape output = Conv2D(name + '.Input', n_channels, DIM, 3, output, he_init=False, cpu=CPU) #print output.shape output = ResidualBlock(name + '.Res1', DIM, 2 * DIM, 3, output, is_training, mode, resample='down') #print output.shape output = ResidualBlock(name + '.Res2', 2 * DIM, 4 * DIM, 3, output, is_training, mode, resample='down') #print output.shape output = ResidualBlock(name + '.Res3', 4 * DIM, 8 * DIM, 3, output, is_training, mode, resample='down') #print output.shape #output = ResidualBlock(name + '.Res4', 8*DIM, 8*DIM, 3, output, is_training, mode, resample='down') output = tf.reshape(output, [-1, 4 * 4 * 8 * DIM]) #print output.shape output = Linear(name + '.Output', 4 * 4 * 8 * DIM, latent_dim, output) #print output.shape return output
def ConvMeanPool(name, input_dim, output_dim, filter_size, inputs, he_init=True, biases=True): output = Conv2D(name, input_dim, output_dim, filter_size, inputs, he_init=he_init, biases=biases, cpu=CPU) output = tf.add_n([ output[:, :, ::2, ::2], output[:, :, 1::2, ::2], output[:, :, ::2, 1::2], output[:, :, 1::2, 1::2] ]) / 4. return output
def high_capacity_decoder(name, z, n_channels, is_training, mode=None, nonlinearity=tf.nn.relu): output = Linear(name + '.Input', z.get_shape().as_list()[1], DIM*4*8*8, z) output = Normalize(name + '.BN0', [0], output, is_training, mode) output = nonlinearity(output) output = tf.reshape(output, [-1, DIM*4, 8, 8]) output = Deconv2D(name + '.1', DIM*4, DIM*4, 5, output) output = Normalize(name + '.BN1', [0,2,3], output, is_training, mode) output = nonlinearity(output) output = Deconv2D(name + '.2', DIM*4, DIM*2, 5, output) output = Normalize(name + '.BN2', [0,2,3], output, is_training, mode) output = nonlinearity(output) output = Deconv2D(name + '.3', DIM*2, DIM // 2, 5, output) output = Normalize(name + '.BN3', [0,2,3], output, is_training, mode) output = nonlinearity(output) output = Conv2D(name + '.4', DIM // 2, n_channels, 5, output) output = tf.reshape(output, [-1, n_channels*DIM*DIM]) return output