Clevert, Djork-Arné, Thomas Unterthiner, and Sepp Hochreiter. "Fast and accurate deep network learning by exponential linear units (elus)." arXiv preprint arXiv:1511.07289 (2015).
- Please install jupyter notebook if not already present.
- Run 'CSE 569 Exponential Linear Unit.ipynb'.
- Different activation functions are written in different blocks. Run the particular block and change the line in the 'model' method (Comment can be found in the method) to run a particular network.
- Cross-Entropy Graph can be recreated by running all 4 networks.
- ~96 % accuracy with a six-layer network (ELU) on the MNIST dataset.