12 (M) 9th February, Friday

I have now finished the implementation of the attacks in Tensorflow, but looking at my model I'm not sure it's quiet right, and I seem to have more trainable variables than I need. Inspecting the size of the layers, I think the shape of the network is a little off so I'll continue to investigate and try to fix it.

Meeting - discussed starting the write up and the new direction of the project. Goals for the next week are to make the motivation and goals of the project clear.