Create and design Generator Model

**You have to define a Generator function and decide its parameters **

We have planned on making a U-Net architecture for the (Generator:https://lmb.informatik.uni-freiburg.de/people/ronneber/u-net/)

Following are the points you have to keep in mind :
- Make Encoder using Convolution layers
- Make Decoder using Transposed Convolution layers
- Create skip connections between the encoder and decoder (as in U-Net).

Start by reviewing the linked U-Net reference and the repository's existing notebook structure. Define the Generator's parameters, then implement an encoder with convolution layers, a decoder with transposed convolutions, and U-Net-style skip connections; it is done when the model design and implementation satisfy all three listed requirements.

Written by the indexing model from the issue text.