Hugging Face's logo

AlreadyExists
/
detr-resnet-50_finetuned_bbatch

Object Detection
Transformers
PyTorch
TensorBoard
detr
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
1
AlreadyExists commited on
Commit
15fdef8
·
1 Parent(s): 76405f8

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +50 -49
README.md CHANGED
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [AlreadyExists/detr-resnet-50_finetuned_bbatch](https://huggingface.co/AlreadyExists/detr-resnet-50_finetuned_bbatch) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 2.2683
18
 
19
  ## Model description
20
 
@@ -40,59 +40,60 @@ The following hyperparameters were used during training:
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
  - num_epochs: 200
 
43
 
44
  ### Training results
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:------:|:----:|:---------------:|
48
- | 2.2996 | 4.17 | 50 | 2.4704 |
49
- | 2.2613 | 8.33 | 100 | 2.4832 |
50
- | 2.2411 | 12.5 | 150 | 2.3990 |
51
- | 2.2071 | 16.67 | 200 | 2.5183 |
52
- | 2.1336 | 20.83 | 250 | 2.3732 |
53
- | 2.103 | 25.0 | 300 | 2.4693 |
54
- | 2.0599 | 29.17 | 350 | 2.3857 |
55
- | 2.0156 | 33.33 | 400 | 2.3739 |
56
- | 2.0183 | 37.5 | 450 | 2.3251 |
57
- | 1.9613 | 41.67 | 500 | 2.3155 |
58
- | 1.9438 | 45.83 | 550 | 2.3530 |
59
- | 1.9139 | 50.0 | 600 | 2.3394 |
60
- | 1.8898 | 54.17 | 650 | 2.3740 |
61
- | 1.8723 | 58.33 | 700 | 2.2917 |
62
- | 1.8438 | 62.5 | 750 | 2.3187 |
63
- | 1.8189 | 66.67 | 800 | 2.3081 |
64
- | 1.8214 | 70.83 | 850 | 2.3503 |
65
- | 1.8046 | 75.0 | 900 | 2.2672 |
66
- | 1.7973 | 79.17 | 950 | 2.2679 |
67
- | 1.7695 | 83.33 | 1000 | 2.3122 |
68
- | 1.7381 | 87.5 | 1050 | 2.2896 |
69
- | 1.7152 | 91.67 | 1100 | 2.2836 |
70
- | 1.6888 | 95.83 | 1150 | 2.3144 |
71
- | 1.6875 | 100.0 | 1200 | 2.2991 |
72
- | 1.6819 | 104.17 | 1250 | 2.3108 |
73
- | 1.6625 | 108.33 | 1300 | 2.3016 |
74
- | 1.6415 | 112.5 | 1350 | 2.3382 |
75
- | 1.6294 | 116.67 | 1400 | 2.2820 |
76
- | 1.6238 | 120.83 | 1450 | 2.3006 |
77
- | 1.6027 | 125.0 | 1500 | 2.3165 |
78
- | 1.6052 | 129.17 | 1550 | 2.2576 |
79
- | 1.5833 | 133.33 | 1600 | 2.2584 |
80
- | 1.5791 | 137.5 | 1650 | 2.2506 |
81
- | 1.5577 | 141.67 | 1700 | 2.3482 |
82
- | 1.5555 | 145.83 | 1750 | 2.3243 |
83
- | 1.5434 | 150.0 | 1800 | 2.2240 |
84
- | 1.5466 | 154.17 | 1850 | 2.2788 |
85
- | 1.5271 | 158.33 | 1900 | 2.3037 |
86
- | 1.5186 | 162.5 | 1950 | 2.3117 |
87
- | 1.5179 | 166.67 | 2000 | 2.2370 |
88
- | 1.5165 | 170.83 | 2050 | 2.2416 |
89
- | 1.5091 | 175.0 | 2100 | 2.2351 |
90
- | 1.4988 | 179.17 | 2150 | 2.2570 |
91
- | 1.5146 | 183.33 | 2200 | 2.2764 |
92
- | 1.501 | 187.5 | 2250 | 2.3092 |
93
- | 1.5014 | 191.67 | 2300 | 2.3145 |
94
- | 1.4909 | 195.83 | 2350 | 2.2830 |
95
- | 1.4868 | 200.0 | 2400 | 2.2683 |
96
 
97
 
98
  ### Framework versions
 
14
 
15
  This model is a fine-tuned version of [AlreadyExists/detr-resnet-50_finetuned_bbatch](https://huggingface.co/AlreadyExists/detr-resnet-50_finetuned_bbatch) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 2.6508
18
 
19
  ## Model description
20
 
 
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
  - num_epochs: 200
43
+ - mixed_precision_training: Native AMP
44
 
45
  ### Training results
46
 
47
  | Training Loss | Epoch | Step | Validation Loss |
48
  |:-------------:|:------:|:----:|:---------------:|
49
+ | 1.1643 | 4.17 | 50 | 2.4138 |
50
+ | 1.1684 | 8.33 | 100 | 2.4418 |
51
+ | 1.1564 | 12.5 | 150 | 2.3408 |
52
+ | 1.1355 | 16.67 | 200 | 2.3903 |
53
+ | 1.1388 | 20.83 | 250 | 2.3484 |
54
+ | 1.139 | 25.0 | 300 | 2.4016 |
55
+ | 1.1098 | 29.17 | 350 | 2.4911 |
56
+ | 1.0869 | 33.33 | 400 | 2.3021 |
57
+ | 1.0871 | 37.5 | 450 | 2.4474 |
58
+ | 1.0793 | 41.67 | 500 | 2.4549 |
59
+ | 1.0691 | 45.83 | 550 | 2.5207 |
60
+ | 1.0539 | 50.0 | 600 | 2.4158 |
61
+ | 1.0579 | 54.17 | 650 | 2.4542 |
62
+ | 1.0365 | 58.33 | 700 | 2.4569 |
63
+ | 1.0221 | 62.5 | 750 | 2.5253 |
64
+ | 1.0173 | 66.67 | 800 | 2.4495 |
65
+ | 1.0241 | 70.83 | 850 | 2.4273 |
66
+ | 1.0111 | 75.0 | 900 | 2.4554 |
67
+ | 1.0127 | 79.17 | 950 | 2.4211 |
68
+ | 0.998 | 83.33 | 1000 | 2.5111 |
69
+ | 0.9869 | 87.5 | 1050 | 2.4077 |
70
+ | 0.9783 | 91.67 | 1100 | 2.5871 |
71
+ | 0.9802 | 95.83 | 1150 | 2.5365 |
72
+ | 0.983 | 100.0 | 1200 | 2.5527 |
73
+ | 0.9719 | 104.17 | 1250 | 2.5728 |
74
+ | 0.9578 | 108.33 | 1300 | 2.5637 |
75
+ | 0.9459 | 112.5 | 1350 | 2.5525 |
76
+ | 0.9353 | 116.67 | 1400 | 2.5476 |
77
+ | 0.94 | 120.83 | 1450 | 2.5374 |
78
+ | 0.9313 | 125.0 | 1500 | 2.6336 |
79
+ | 0.9236 | 129.17 | 1550 | 2.5556 |
80
+ | 0.9129 | 133.33 | 1600 | 2.5768 |
81
+ | 0.9231 | 137.5 | 1650 | 2.5904 |
82
+ | 0.9093 | 141.67 | 1700 | 2.6503 |
83
+ | 0.9169 | 145.83 | 1750 | 2.6057 |
84
+ | 0.9082 | 150.0 | 1800 | 2.6561 |
85
+ | 0.911 | 154.17 | 1850 | 2.6234 |
86
+ | 0.9019 | 158.33 | 1900 | 2.6442 |
87
+ | 0.8892 | 162.5 | 1950 | 2.6090 |
88
+ | 0.8891 | 166.67 | 2000 | 2.5849 |
89
+ | 0.8898 | 170.83 | 2050 | 2.6186 |
90
+ | 0.8745 | 175.0 | 2100 | 2.7664 |
91
+ | 0.8714 | 179.17 | 2150 | 2.6261 |
92
+ | 0.8902 | 183.33 | 2200 | 2.6510 |
93
+ | 0.8781 | 187.5 | 2250 | 2.7035 |
94
+ | 0.8826 | 191.67 | 2300 | 2.5627 |
95
+ | 0.8733 | 195.83 | 2350 | 2.5455 |
96
+ | 0.8694 | 200.0 | 2400 | 2.6508 |
97
 
98
 
99
  ### Framework versions