Corresponding to http://ieeexplore.ieee.org/abstract/document/7377956/
- Strictly based on MLP and the code from http://deeplearning.net/.
- Circuits are equally divided.
- Outside dropout rate and sparsity coefficient, other circuit descriptions are similar
- Weight/Learning rate decay is not included
- execute.py : experiment configuration for mass execution
- model.py : PC definition
- layer.py : layer defintion