Hugging Face's logo

MedicalVision
/
detr_nih_10ep

Object Detection
Transformers
Safetensors
detr
Model card Files
xet
Community
npvinHnivqn commited on
Commit
c10b33e
·
verified ·
1 Parent(s): 4e7c9e3

Update README file

Browse files
Files changed (1) hide show
  1. README.md +46 -12
README.md CHANGED
@@ -12,31 +12,65 @@ IoU metric: bbox
12
  Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
13
  Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
14
  Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.000
15
- Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.000
16
- Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.003
17
- Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.013
18
  Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
19
  Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
20
- Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.013
21
  ```
 
22
  ## After training result
23
  ```
24
  IoU metric: bbox
25
- Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.003
26
- Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.007
27
  Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.002
28
  Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
29
  Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
30
- Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.003
31
- Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.046
32
- Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.085
33
- Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.098
34
  Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
35
- Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
36
- Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.101
37
  ```
 
38
  ## Config
39
  - dataset: NIH
40
  - original model: facebook/detr-resnet-50
41
  - lr: 0.0001
42
  - max_epochs: 10
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12
  Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
13
  Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
14
  Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.000
15
+ Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.001
16
+ Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.004
17
+ Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.011
18
  Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
19
  Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
20
+ Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.012
21
  ```
22
+
23
  ## After training result
24
  ```
25
  IoU metric: bbox
26
+ Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.004
27
+ Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.010
28
  Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.002
29
  Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
30
  Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
31
+ Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.004
32
+ Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.042
33
+ Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.079
34
+ Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.093
35
  Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
36
+ Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.007
37
+ Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.094
38
  ```
39
+
40
  ## Config
41
  - dataset: NIH
42
  - original model: facebook/detr-resnet-50
43
  - lr: 0.0001
44
  - max_epochs: 10
45
+
46
+ ## Logging
47
+
48
+ ### Training process
49
+ ```
50
+
51
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
52
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
53
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
54
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
55
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
56
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
57
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
58
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
59
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
60
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
61
+ ```
62
+
63
+ ### Validation process
64
+ ```
65
+ {'validation_loss': tensor(6.4491, device='cuda:0'), 'validation_loss_ce': tensor(1.6310, device='cuda:0'), 'validation_loss_bbox': tensor(0.5587, device='cuda:0'), 'validation_loss_giou': tensor(1.0123, device='cuda:0'), 'validation_cardinality_error': tensor(8.9375, device='cuda:0')}
66
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
67
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
68
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
69
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
70
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
71
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
72
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
73
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
74
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
75
+ {'training_loss': tensor(2.1957, device='cuda:0'), 'train_loss_ce': tensor(0.4394, device='cuda:0'), 'train_loss_bbox': tensor(0.1586, device='cuda:0'), 'train_loss_giou': tensor(0.4816, device='cuda:0'), 'train_cardinality_error': tensor(1.0625, device='cuda:0'), 'validation_loss': tensor(2.2341, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.1467, device='cuda:0'), 'validation_loss_giou': tensor(0.5182, device='cuda:0'), 'validation_cardinality_error': tensor(1.1295, device='cuda:0')}
76
+ ```