Simulate collaborative ML scenarios, experiment multi-partner learning approaches and measure respective contributions of different datasets to model performance.
The RMSprop optimizer can be changed to adam with a different learning rate. It greatly improves the training performance, as shown below.
However, even if I tried a few learning rates and decay values, these are propably not the perfect hyperparameters, do not hesitate to suggest another setup.
The RMSprop optimizer can be changed to adam with a different learning rate. It greatly improves the training performance, as shown below.
However, even if I tried a few learning rates and decay values, these are propably not the perfect hyperparameters, do not hesitate to suggest another setup.
Signed-off-by: arthurPignet arthur.pignet@mines-paristech.fr