perf: recalibrate vram_plan (leave GPU headroom) + refresh perf README

Fix model-intrinsic params in the model (input channels / class count / network architecture) instead of exposing them in the config

app.json: drop redundant top-level patch_size (now read from Prediction.yml)

Update model card, app.json (patch size + VRAM plan) and inference configs

Make SAM_Perceptual metric explicit (train=False) in evaluation configs

Use SAM_Perceptual(train=True, weights=[0,1,1,0]) as training loss instead of IMPACTReg

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/Uncertainty.yml

Update MR/Uncertainty.yml

Update MR/Prediction.yml

Update MR/Evaluation.yml

Update MR/Evaluation.yml

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/Prediction.yml

Update whole-body sCT model

Update whole-body sCT model

Update MR/Prediction.yml

Update MR/app.json

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/app.json

Update MR/app.json

Update MR/app.json

Update MR/Uncertainty.yml

Update MR/Model.py

Update MR/Evaluation.yml

Update MR/requirements.txt

Update MR/Prediction.yml

Rename MR/metadata.json to MR/app.json

Update MR/Uncertainty.yml

Update MR/Prediction.yml

Update MR/Evaluation.yml

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/Prediction.yml

Update MR/requirements.txt

Update MR/requirements.txt

Update MR/Uncertainty.yml

Update MR/Prediction.yml

Update MR/Evaluation.yml

Update MR/Evaluation.yml

Update MR/Uncertainty.yml