This code uses a Conv-Va AE model to reconstruct images. Many codes have been reviewed and good points have been added to prepare a good sample for those who are trying to learn autoencoders specialy for images. !! there is no new idea or contribution (the stucture is inspireed from various papers and codes and this code just aims to be a good sample for those whom are trying to use Conv-va-AE in their projects)