Write the rest missing layers like classification, softmax and regression layers. Write the main CNN class, that holds the layers, its training process and the classify method. In this point we should be able to train some networks and test their functionality. Once again, check its compatibility with the Matlab toolbox.
Through the whole process I plan to keep the code well documented using Doxygen. In case we use the Tensorflow's Python interface, we could look for another way of doing it (like Sphinx).
Phase 3, until 29 August:
Add extra features like the activations class, that can display a layer activations, and the possibility to load pretrainned networks. We could start by defining the AlexNET and VGG16 networks. This could be extended with an exporting/importing tool to process CNNs' architecture and weights saved in files.
[[Category: Summer of Code]]