Appfl

New Features

- Support batched MPI, with documentation available [here](https://appfl.ai/en/latest/tutorials/examples_batched_mpi.html).
- Add more data readiness metrics such as PCA plot in PR 208
- Backend support for [service.appfl.ai](https://appflx.link/).
- Add documentation for service.appfl.ai at [here](https://appfl.ai/en/latest/tutorials/appflx/index.html).
- Add logging capabilities to the server side to log the training metadata such as the training and validation losses.
- Change documentation theme to `furo`.

Community Standards
- Add [pull request template](https://github.com/APPFL/APPFL/blob/main/.github/pull_request_template.md) and issue templates
- Add [contribution guidance](https://appfl.ai/en/latest/contribution/index.html)
- Add dependabot for auto github action version check

New Features

- Add the feature to generate data readiness reports on all client data by from kaveenh (PR 202)

- Update the documentation for adding custom action at [here](https://appfl.ai/en/latest/tutorials/examples_custom_action.html).

New Features
- Add documentation for using APPFL with Globus Compute for secure distributed training at [here](https://appfl.ai/en/latest/tutorials/examples_globus_compute.html).

Bug Fixes
- Fix an issue with Globus Compute at this [commit](https://github.com/APPFL/APPFL/commit/705b5af64389c77e1c0f9f21d1d86c0cc33cd067).

New Features
- Add trackback information to the gRPC server to help debug the server-side errors.
- Add a video tutorials for [installing APPFL on AWS](https://youtu.be/ihPofoQwUMs), [creating SSL-encrypted gRPC server](https://youtu.be/3n8a026VqdQ), and [using APPFL to finetune a ViT](https://youtu.be/m4rdOub2Y_o).

Bug Fixes
- Handle corner cases for server aggregators when the keys in client local models are not consistent with the global model keys.

New Features

- Add a new command line interface (CLI), `appfl-setup-ssl`, to create necessary certificates for creating SSL-secured gRPC connections between the server and clients.
- Add a tutorial on how to use the CLI, `appfl-setup-ssl` to create certificates for the server and clients, and enable SSL-secured gRPC connections between the server and clients at [here](https://appfl.ai/en/latest/tutorials/examples_ssl.html).
- Add a detailed step-by-step tutorial on how to define custom action with an example to generate a data readiness report on all client data at [here](https://appfl.ai/en/latest/tutorials/examples_custom_action.html).
- Add a APPFL [YouTube channel](https://www.youtube.com/channel/UCzwiJboiJW3dLI0UndnDy5g) to provide video tutorials on how to use APPFL for federated learning research in the future.

Bug Fixes
Fix the [issue](https://github.com/APPFL/APPFL/issues/197) regarding client gradient clipping. The clipping is now applied before weights update.

New Features

- For the aggregators, the model architecture is set to be an optional initialization parameter, and the aggregators only aggregate the parameters sent by the clients instead of the whole set of model parameters. This is useful when doing federated fine-tuning or federated transfer learning where only part of model parameters are updated / the model architecture is unknown to the aggregator.

- Support easy integration of custom trainer/aggregator: user only needs to provide the custom trainer/aggregator class name and the path to the definition file in the configuration file to use it, instead of modifying the source code.

- Add a detailed step-by-step tutorial on how to use APPFL to fine-tune a ViT model with a custom trainer at https://appfl.ai/en/latest/tutorials/examples_vit_finetuning.html