feat: no VRAM plan, the prediction config asks for a measured batch

The apps declare no `vram_plan` any more and their `Prediction.yml` says `batch_size: 0`: konfai measures the batch its forward fits on the GPU (one patch at a time on a CPU). Needs the konfai release that measures the batch: an older konfai refuses a batch of 0.

README.md

@@ -37,7 +37,7 @@ impact-seg-konfai segment body -i input.nii.gz -o output/
 
 ## ⚡ Performance & VRAM
 
-Benchmarked on a single **NVIDIA RTX PRO 5000 (24 GB)** with a real whole-body CT (295 × 259 × 219, 2 mm). The batch size is **auto-selected from your free GPU VRAM**.
+Benchmarked on a single **NVIDIA RTX PRO 5000 (24 GB)** with a real whole-body CT (295 × 259 × 219, 2 mm). The batch size is **measured on your GPU** (a forward of one patch, then of two, then the batch the free VRAM holds).
 
 | Free VRAM | Batch (auto) | Peak VRAM | Time / case |
 |:--|:--|:--|:--|

body/Prediction.yml

@@ -38,7 +38,7 @@ Predictor:
     subset: None
     dataset_filenames:
     - ./Dataset/:mha
-    batch_size: 32
+    batch_size: 0
     num_workers: None
     pin_memory: false
     prefetch_factor: None

body/app.json

@@ -83,31 +83,5 @@
             "color": "#43AA8B"
         }
     },
-    "vram_plan": {
-        "8": {
-            "patch_size": [
-                1,
-                192,
-                192
-            ],
-            "batch_size": 160
-        },
-        "16": {
-            "patch_size": [
-                1,
-                192,
-                192
-            ],
-            "batch_size": 320
-        },
-        "24": {
-            "patch_size": [
-                1,
-                192,
-                192
-            ],
-            "batch_size": 512
-        }
-    },
     "task": "segmentation"
 }