A model previously trained in the Tabular_Data example is used to demonstrate
transfer learning capabilities. A local PyTorch model (.pth file) created using
the torch.nn.Sequential module is uploaded to TripleBlind system to start the
process.

From here the 2_train_train.py begins additional training against previously
uploaded datasets to further train the model.  Once training is complete, the
refined model is downloaded for use in future transfer trainings.  Each
transfer training will create a new asset to retain model versions.

The final scripts run inference against local and remote models.
