Javier

@abulka, thanks for the above. Did you manage to pack the application? I manage to compile it ok, but still, have the same error when running the generated executable(MODULE_VERSION...). I...

this is because two independent simulations aren't aware of the other simulation running... to fix this you need to pass the resources you want to use when calling `ray.init()` (which...

Hi @nikita-kotsehub and @sisco0. Thanks for looking into this. I agree adding support for Debian is important since it remains the recommended OS for RPi. I've created a new PR...

I think you are right. we can see that the network learns because the batchnorm parameters do change. But that, of course, is not enough to reach high accuracies....

Hi. Probably you are using a different version of TensorBoardX. You can solve this error by updating line 19 of `search.py` by defining the input argument as `logdir` instead of...

> Please elaborate your question a bit more, what type of model do you want to federate with Flower? I think @ouaelesi was wondering if Flower supports other forms of...

Hi @FrancescoPase, I made some changes to your baseline. They are in `baselines/fedpm` (as opposed to your previous commits in `baselines/FedPM`). Take a look at the environment setup section in...

We have re-structured the baselines and some of these ideas have been incorporated. Hydra is here to stay!

Hey @prathapkumarbaratam , you can find both `FedProx` and `FedNova` in the `baselines/` directory: * `FedNova`: https://github.com/adap/flower/tree/main/baselines/fednova * `FedProx`: https://github.com/adap/flower/tree/main/baselines/fedprox Many more baselines are easily accessible from our docs: https://flower.dev/docs/baselines/

Closing this since the `embedded_devices` example was updated recently, adding quite a few simplifications and making it more versatile overall. Please feel free to open a new issue if you...