Python Deep Learning Projects

Defining the autoencoder

The convolution autoencoder we are going to build will accept 14*14*1 images as input with 28*28*1 images as the targets, and will have the following characteristics:

In the encoder:

• The first layer is a convolution 2-D layer with 64 filters of size 3*3, followed by batch normalization, with activation as relu, followed by downsampling with MaxPooling2D of size 2*2
• The second layer, or the final layer in this encoder part, is again a convolution 2-D layer with 128 filters of size 3*3, batch normalization, with activation as relu

In the decoder:

• The first layer is a convolution 2-D layer with 128 filters of size 3*3 with activation as relu, followed by upsampling that's performed with UpSampling2D 
• The second layer ...