Add Flir baseline implementation

[ccll1122]

Baseline: Add FLIR federated learning implementation

Issue

Description

I am contributing a new baseline implementation of FLIR CVPR-2022] Federated Class-Incremental Learning to the Flower framework. This baseline implements federated learning experiments from our paper (paper citation).

[jafermarq]

Hi @ccll1122 and @Luosmallrui, thanks for creating the PR to include FLIR in the Flower Baselines collection!! Before FLIR can be added, you'll need to adjust how the code is organized. Please follow the instructions on how to add a new baseline. In short, these are the steps:

1. From a new python environment: pip install -U flwr
2. Create the baseline structure using flwr new:
   • Clone the flower repo
   • Cd to directory where all baselines are
   • Create a baselines skeleton project: flwr new flir (then follow the prompt)
3. The final step above would have created the flir directory and some basic code inside, please adjust it with your own code but without making structural changes (i.e. the pyproject.toml needs to be there, but you can edit, you need to complete the indicated areas in the generated README.md, your baseline should be run with flwr run).

You can use as reference for how to structure the code some of the examples in flower, for example https://github.com/adap/flower/tree/main/examples/advanced-pytorch.

Please let me know if you need some help!