This is the repo for the second research paper in FYS-STK 3155 Applied data analysis and Machine Learning. It concerns numerical optimization techniques and feed forward neural networks.
Gregor Kajda \ Eric Emanuel Reber \ Jonatan Hoffmann Hanssen
The paper can be found at doc/paper.pdf. The implementations of gradient descent, different schedulers and the FFNN can be found at src/Schedulers.py, src/FFNN.py and src/utils.py. The rest of the src directory contains scripts used to create various plots in the report. A description of which script is used to create which script can be found in the appendix of the paper.