Hugging Face's logo

VBoussot
/
ImpactSynth

Image-to-Image
konfai
medical-imaging
synthetic-ct
image-synthesis
radiotherapy
mri
cbct
totalsynth
Model card Files
xet
 Community
VBoussot commited on
Commit
8006631
·
verified ·
1 Parent(s): a07f576

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

Browse files

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.

Files changed (8) hide show
  1. CBCT/Prediction.yml +1 -1
  2. CBCT/app.json +0 -26
  3. Finetune/Prediction.yml +1 -1
  4. Finetune/app.json +0 -26
  5. MR/Prediction.yml +1 -1
  6. MR/app.json +0 -26
  7. MR_CBCT/Prediction.yml +1 -1
  8. MR_CBCT/app.json +0 -26
CBCT/Prediction.yml CHANGED
@@ -84,7 +84,7 @@ Predictor:
84
  subset: None
85
  dataset_filenames:
86
  - ./Dataset:mha
87
- batch_size: 8
88
  outputs_dataset:
89
  Tanh:
90
  OutputDataset:
 
84
  subset: None
85
  dataset_filenames:
86
  - ./Dataset:mha
87
+ batch_size: 0
88
  outputs_dataset:
89
  Tanh:
90
  OutputDataset:
CBCT/app.json CHANGED
@@ -41,31 +41,5 @@
41
  }
42
  }
43
  },
44
- "vram_plan": {
45
- "8": {
46
- "patch_size": [
47
- 1,
48
- 512,
49
- 512
50
- ],
51
- "batch_size": 16
52
- },
53
- "16": {
54
- "patch_size": [
55
- 1,
56
- 512,
57
- 512
58
- ],
59
- "batch_size": 28
60
- },
61
- "24": {
62
- "patch_size": [
63
- 1,
64
- 512,
65
- 512
66
- ],
67
- "batch_size": 48
68
- }
69
- },
70
  "task": "synthesis"
71
  }
 
41
  }
42
  }
43
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
44
  "task": "synthesis"
45
  }
Finetune/Prediction.yml CHANGED
@@ -84,7 +84,7 @@ Predictor:
84
  subset: None
85
  dataset_filenames:
86
  - ./Dataset:mha
87
- batch_size: 8
88
  outputs_dataset:
89
  Tanh:
90
  OutputDataset:
 
84
  subset: None
85
  dataset_filenames:
86
  - ./Dataset:mha
87
+ batch_size: 0
88
  outputs_dataset:
89
  Tanh:
90
  OutputDataset:
Finetune/app.json CHANGED
@@ -41,31 +41,5 @@
41
  }
42
  }
43
  },
44
- "vram_plan": {
45
- "8": {
46
- "patch_size": [
47
- 1,
48
- 512,
49
- 512
50
- ],
51
- "batch_size": 16
52
- },
53
- "16": {
54
- "patch_size": [
55
- 1,
56
- 512,
57
- 512
58
- ],
59
- "batch_size": 28
60
- },
61
- "24": {
62
- "patch_size": [
63
- 1,
64
- 512,
65
- 512
66
- ],
67
- "batch_size": 48
68
- }
69
- },
70
  "task": "synthesis"
71
  }
 
41
  }
42
  }
43
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
44
  "task": "synthesis"
45
  }
MR/Prediction.yml CHANGED
@@ -84,7 +84,7 @@ Predictor:
84
  subset: None
85
  dataset_filenames:
86
  - ./Dataset:mha
87
- batch_size: 8
88
  outputs_dataset:
89
  Tanh:
90
  OutputDataset:
 
84
  subset: None
85
  dataset_filenames:
86
  - ./Dataset:mha
87
+ batch_size: 0
88
  outputs_dataset:
89
  Tanh:
90
  OutputDataset:
MR/app.json CHANGED
@@ -41,31 +41,5 @@
41
  }
42
  }
43
  },
44
- "vram_plan": {
45
- "8": {
46
- "patch_size": [
47
- 1,
48
- 512,
49
- 512
50
- ],
51
- "batch_size": 16
52
- },
53
- "16": {
54
- "patch_size": [
55
- 1,
56
- 512,
57
- 512
58
- ],
59
- "batch_size": 28
60
- },
61
- "24": {
62
- "patch_size": [
63
- 1,
64
- 512,
65
- 512
66
- ],
67
- "batch_size": 32
68
- }
69
- },
70
  "task": "synthesis"
71
  }
 
41
  }
42
  }
43
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
44
  "task": "synthesis"
45
  }
MR_CBCT/Prediction.yml CHANGED
@@ -84,7 +84,7 @@ Predictor:
84
  subset: None
85
  dataset_filenames:
86
  - ./Dataset:mha
87
- batch_size: 8
88
  outputs_dataset:
89
  Tanh:
90
  OutputDataset:
 
84
  subset: None
85
  dataset_filenames:
86
  - ./Dataset:mha
87
+ batch_size: 0
88
  outputs_dataset:
89
  Tanh:
90
  OutputDataset:
MR_CBCT/app.json CHANGED
@@ -41,31 +41,5 @@
41
  }
42
  }
43
  },
44
- "vram_plan": {
45
- "8": {
46
- "patch_size": [
47
- 1,
48
- 512,
49
- 512
50
- ],
51
- "batch_size": 16
52
- },
53
- "16": {
54
- "patch_size": [
55
- 1,
56
- 512,
57
- 512
58
- ],
59
- "batch_size": 28
60
- },
61
- "24": {
62
- "patch_size": [
63
- 1,
64
- 512,
65
- 512
66
- ],
67
- "batch_size": 48
68
- }
69
- },
70
  "task": "synthesis"
71
  }
 
41
  }
42
  }
43
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
44
  "task": "synthesis"
45
  }