Refactoring DataLoading and Training
Problem to solve
As AQ-Bench user, I want to have separate modules for training and inference.
Machine Learning Models and Training
- The folder "models" would contain the definitions of the neural network and random forest models.
- The training of the model to be executed through the run_train.py file.
- At the end of the training, the model has to be saved.
- To run the inference the saved models need to be loaded.
What does success look like?
- There are clear definitions for the neural network and random forest models.
- The training and inference can be executed independently.