PyTorch Versions of TC1 and P1B2 Benchmarks

[knangia04]

• Code is generic and handles the neural network model as defined in the model_pytorch.txt file (similar to the TensorFlow/Keras version).
• baseline_pytorch.py is the main function of the benchmark.
• Updated pytorch_utils.py to add additional functionality.
• torch_deps folder consists of dependencies that are required for the PyTorch benchmark, including a model class, dataset DataLoader class, classification network, training, validation, and test functions, and weight initialization.

[jmohdyusof]

How does this commit relate to #95 which seems to be similar (at least in name)