Structure of files:

Files with code - my_utils.py: has all the util files needed for the training in general - basic_steps_one.py: code example for applying the concept of SL operation with one split -- it shows just the basic operations between one client a server - basic_steps_two.py: code example for applying the concept of SL operation with two splits -- it shows just the basic operations between one client a server - /application: folder that contain an example implementation of a whole framework for SplitFed -- it does not contain networking - /application/model.py: implement a class for model (simple-cnn) that returns a model part of a model - /application/train.py: main code, i.e., server side - /application/client.py: client side, note that this could be a standalone thread/process

Files with notes: - requirements.txt: important libraries with links - set_up_rpi.txt: notes on how to set-up RPI - set_up_jtson.txt: notes on how to set-up Jtson - git_list.txt: list of github repos

Other notes:

Dataloaders and partition: In order to build data loaders we need to define the type of data partition. For the following experiments we use IID, but in the my_utilis.py file, the generate_partitioner() function contains other types of partiotions.