search

diaoenmao / HeteroFL-Computation-and-Communication-Efficient-Federated-Learning-for-Heterogeneous-Clients

[ICLR 2021] HeteroFL: Computation and Communication Efficient Federated Learning for Heterogeneous Clients
MIT License
150 stars 33 forks source link

Model training on custom dataset #10

Closed chillum-codeX closed 1 year ago

chillum-codeX commented 1 year ago

Hey I wants to do a custom model training For example taking covid, Ctscan and Mri data Can you please guide me

diaoenmao commented 1 year ago

I think you can write your own dataset files like the ones in ./dataset/mnist.py, ./dataset/cifar.py. A more clear pipeline I used can be found in RPipe.

chillum-codeX commented 1 year ago

Hey thanks for reply can you provide an example with COVID Images dataset how can we train the moel on covid images and CT MRI dataset please proovie a running script , it would be great and helpful for me

diaoenmao commented 1 year ago

Sorry. I don't have access to these resources.

chillum-codeX commented 1 year ago

Hey this is the resource link for data i need an working idea rest i will do https://github.com/ml-workgroup/covid-19-image-repository

diaoenmao commented 1 year ago

Sorry. I think this is beyond the scope of this repository. The source code only provides support for datasets used in the paper.