having trouble reconstructing the exact input using a variational Autoencoder

Question

I\'m trying to implement a variational autoencoder for my research. I\'m using a dataset that has 14 features and I\'m trying to learn to reconstruct the exact values for all of