Cats vs Non-cats using Transfer Learning - What are we going to do now?

We are going to build a model to classify cats vs non-cats, using transfer learning.
We are going to build it on top of the pre-trained VGG16 models.
We shall use the weights of the model, and fine-tune it by adding some layers on top of that, in order to customize it for our own purpose - cat vs non-cat classification.
We use the same weights of the CNNs by freezing them, and we remove the top(which are dense layers) to add our customized dense layers. The customized top dense layers are trained(which is known as fine-tuning), to make them especially learn about our data set.

enter image description here