distributed-learning-contributivity
distributed-learning-contributivity copied to clipboard
LabeliaLabs
Simulate collaborative ML scenarios, experiment multi-partner learning approaches and measure respective contributions of different datasets to model performance.
The batch size is clipped here --> https://github.com/SubstraFoundation/distributed-learning-contributivity/blob/b72fa98c0b4db45d368f577d0f6d1a861b1610c2/scenario.py#L584 So if it's clipped it means that sometimes we don't use all the data in the dataset. But we don't give any...
The MNIST nn have 1,199,882 parameters, for an accuracy of 0.9893 +/- 0.01 (keras fit, epochs=12, batch_size=168) To compare with, the cifar model have 1,250,858 parameters, and reach 0.7888 +/-...
See above: One example, sometime there is #, sometime there isn't. 2021-01-14 17:54:30 | INFO | ### Splitting data among partners: 2021-01-14 17:54:30 | INFO | Train data split: 2021-01-14...
Currently we can set the validation and test sets as local or global. It could be interesting to had both usages
@bowni @Thomas-Galtier @arthurPignet What would happen if we removed this threshold ? https://github.com/SubstraFoundation/distributed-learning-contributivity/blob/808ee93c8593d3b226d3d1aeaa370c6d9d9689f8/mplc/contributivity.py#L452
The repo is now 1.1GB which is not okay. It is likely that a dataset has been added somewhere. I will investigate this issue, but any help is welcome for...
When using the ESC50 dataset, we preprocess the data at each instanciation of a dataset. However the preprocessing is quite long. It would be great to store the preprocessed inputs...
We are used to having good performance results for MNIST dataset (often reaching >80% accuracy) independently from scenario configuration, which allows for good comparison of contributivity methods implemented For Cifar...
Metadata
Owner
Metadata
Simulate collaborative ML scenarios, experiment multi-partner learning approaches and measure respective contributions of different datasets to model performance.