rayfed
rayfed copied to clipboard
A multiple parties joint, distributed execution engine based on Ray, to help build your own federated learning frameworks in minutes.
Our documentation can still be significantly improved.
The PR #194 skip client mode ut but we should fix it. https://github.com/ray-project/rayfed/actions/runs/7498914983/job/20414798670
- [ ] Address all P0 issues - [ ] Add a change log file - [ ] Finish the document - [ ] release test - [ ] release...
Now that we separate `ray.init` from `fed.init`, there's no way to reach the cluster-level information, since each `fed.init` starts and only starts a job session. Unless there's a global actor...
- [ ] Move grpc_config out of config file. - [ ] Make new communication component easy to plugin.
https://github.com/ray-project/rayfed/blob/main/tests/test_listen_addr.py#L59
- [ ] Remove cluster config concept, use config to replace job_config. - [ ] Introduce config_for_each_party.