Foundation Model-based Cloud-Device Collaborative Online Adaptation with Effective Linking

---

Environment setting

• install conda env

conda env create -f env.yaml

• install pytorch (e.g., we use pytorch version of 2.3.1

conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia

• install mmsegmentation (reference)

pip install -U openmim
mim install mmengine
mim install "mmcv>=2.0.0"

pip install -v -e .

---

Run

Cloud Computer = demo-raeyo@172.27.183.243://home/demo-raeyo/Workspace/cloud-edge-collaborative-adaptation

# Cloud Computer Terminal 1
python host/run_cloud_manager_baseline.py # For baseline
---
python host/run_cloud_manager_adaptformer.py # For adaptformer

# Cloud Computer Terminal 2
sh run_model_trainer.sh # For baseline
---
sh run_model_trainer_adaptformer.sh # For adaptformer

---

Device computer = device@172.27.183.242:/home/device/workspace/cloud-edge-collborative-adaptation

# Device Computer Terminal 1
sh run_client_segb1_project.sh # For baseline

sh run_client_segb1_adapt_project.sh # For adaptformer