Gustavo Bertoli
Gustavo Bertoli
Nice contribution @jay-johnson and I liked the suggestion to create a Jupyter Notebooks section!
I did tests with flwr 0.19.0. If trying to execute two consecutive simulations (fl.simulation.start_simulation), similar error (exhausting computational resources) arise: ``` --------------------------------------------------------------------------- ValueError Traceback (most recent call last) Input In...
@jafermarq I am not running two independent simulation simultaneously, but running one and once it stops trying to run another one. Also, my configuration do not have GPUs. Do you...
@EliasBrohammer the GradientBoostingClassifier does not have the coef_ atribute, looking to the sklearn API, probably the [get_params method](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingClassifier.html#sklearn.ensemble.GradientBoostingClassifier.get_params) may be useful.
> Thanks for the the detailed report @gubertoli, does it change anything if you do a `pip install -U flwr[simulation]` (note the missing `"`)? > > The Flower tutorial installs...
> Did you try changing "num_rounds=NUM_ROUNDS," to "config=fl.server.ServerConfig(num_rounds=NUM_ROUNDS)," ? Good point, I just checked the changes for simulation from `0.19.0` to `1.0.0` [here](https://flower.dev/docs/upgrade-to-flower-1.0.html). However, even complying with the new syntax,...
@danieljanes just tested with Flower 1.1 and the problem still happens: ``` --------------------------------------------------------------------------- ImportError Traceback (most recent call last) Input In [26], in () 4 strategy=fl.server.strategy.FedAvg( 5 #fraction_fit=0.1, # Sample...
I had this same issue with flwr 1.7.0 and Python 3.12 (_trying pip install flwr[simulation] downgrades to flwr==0.16.0_). To solve, I had to downgrade to Python 3.10. Seems to be...
Great initiative @cinience with a chinese version. You will benefit people that does not speak english spreading knowledge. Anyone wants to do the same for Portuguese ?
It seems that `FederatedDataset` class is only downloading the dataset from HF. Probably should be changed to enable a custom dataset in the HF's `dataset` format be directly referenced instead...