Ewen Wang

Tensorflow data loader example

2

The [tensorflow client example ](https://github.com/SymbioticLab/FedScale/blob/master/examples/tensorflow_engine/tf_client.py) is very helpful, but one suggestion for improvement is to add a [tf Dataset](https://www.tensorflow.org/api_docs/python/tf/data/Dataset) based dataloader. I think the existing dataloaders are all using torch,...

Questions about abstracting client communication from simulation

3

Hi, we are interested in building a backend aggregator/coordinator that can eventually support production traffic, using FedScale for simulating device training and traffic. This means we wouldn't be implementing an...

Pull mechanism for client-server communication

7

FedScale is currently built under an assumption that the central server communicates with clients based on push mechanism where the server initiates signals (handshake/training/notTraining/etc) to available devices. Is it possible...

[Async simulation] Implementation idea for task scheduling

1

### Description Hi FedScale team, here's my suggestion on how to implement the async simulation mode using device traces without needing a constant arrival parameter (related to #162): ``` sort...

Improve documentation in various components

1

### Description Some code comments and docstrings should be added, especially for resource_manager.py, data_loader.py, client_manager.py, etc. Also, some model diagram on what each component would be helpful for people new...

Batch normalization layer from Keras only converts to BatchNormalization op if dims >= 4

1

Hi Onnx team, I'm running into a model conversion issue for Kera's BatchNorm layer. The BatchNorm layer is only converted into a BatchNorm Op in Onnx when input Dim>=4. I...