digivi13

to find out which cluster new data belongs to, get the embedded representations for all the images and perform a PCA on the entire set of embedded training data. Then...

To visualize the components in 2d, perform predictions using only the encoder to get the embedded vectors for each sample. Then perform a PCA with two components on this array...

or possibly use model.encoder.predict(X)

make images have a single color channel, scale them down to 28x28. Or you can use larger images and adjust the size of input the input to be equal to...

I got the same error, solved it by adding weights to the base config like so: ` base_config['weights'] = self.initial_weights` When training I save the model to a json file...