Hugging Face's logo

WKLI22
/
detr-resnet-50_finetuned_cppe5

Object Detection
Transformers
PyTorch
TensorBoard
Safetensors
detr
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
WKLI22 commited on
Commit
698570b
·
verified ·
1 Parent(s): 2d39722

Training in progress, step 5

Browse files
Files changed (4) hide show
  1. README.md +71 -50
  2. config.json +1 -1
  3. model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [WKLI22/detr-resnet-50_finetuned_cppe5](https://huggingface.co/WKLI22/detr-resnet-50_finetuned_cppe5) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 0.4986
19
 
20
  ## Model description
21
 
@@ -35,67 +35,88 @@ More information needed
35
 
36
  The following hyperparameters were used during training:
37
  - learning_rate: 1e-05
38
- - train_batch_size: 17
39
- - eval_batch_size: 17
40
  - seed: 42
41
- - gradient_accumulation_steps: 6
42
- - total_train_batch_size: 102
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
- - num_epochs: 3
46
  - mixed_precision_training: Native AMP
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
- | 0.4724 | 0.07 | 2 | 0.5355 |
53
- | 0.4907 | 0.14 | 4 | 0.5277 |
54
- | 0.498 | 0.21 | 6 | 0.5214 |
55
- | 0.534 | 0.28 | 8 | 0.5274 |
56
- | 0.5305 | 0.36 | 10 | 0.5294 |
57
- | 0.5096 | 0.43 | 12 | 0.5273 |
58
- | 0.5149 | 0.5 | 14 | 0.5158 |
59
- | 0.5099 | 0.57 | 16 | 0.5163 |
60
- | 0.5308 | 0.64 | 18 | 0.5217 |
61
- | 0.5164 | 0.71 | 20 | 0.5076 |
62
- | 0.4932 | 0.78 | 22 | 0.5050 |
63
- | 0.5151 | 0.85 | 24 | 0.5102 |
64
- | 0.4982 | 0.92 | 26 | 0.5044 |
65
- | 0.526 | 0.99 | 28 | 0.5096 |
66
- | 0.5034 | 1.07 | 30 | 0.4980 |
67
- | 0.5155 | 1.14 | 32 | 0.5067 |
68
- | 0.513 | 1.21 | 34 | 0.5011 |
69
- | 0.5019 | 1.28 | 36 | 0.5066 |
70
- | 0.4704 | 1.35 | 38 | 0.5094 |
71
- | 0.5404 | 1.42 | 40 | 0.5126 |
72
- | 0.5263 | 1.49 | 42 | 0.5062 |
73
- | 0.4729 | 1.56 | 44 | 0.5223 |
74
- | 0.5032 | 1.63 | 46 | 0.5073 |
75
- | 0.476 | 1.7 | 48 | 0.5111 |
76
- | 0.4823 | 1.78 | 50 | 0.5094 |
77
- | 0.5223 | 1.85 | 52 | 0.5042 |
78
- | 0.4855 | 1.92 | 54 | 0.4962 |
79
- | 0.5038 | 1.99 | 56 | 0.5006 |
80
- | 0.5196 | 2.06 | 58 | 0.5022 |
81
- | 0.4847 | 2.13 | 60 | 0.4943 |
82
- | 0.4697 | 2.2 | 62 | 0.5007 |
83
- | 0.4893 | 2.27 | 64 | 0.5041 |
84
- | 0.4939 | 2.34 | 66 | 0.4910 |
85
- | 0.5093 | 2.41 | 68 | 0.4974 |
86
- | 0.4884 | 2.49 | 70 | 0.4962 |
87
- | 0.5087 | 2.56 | 72 | 0.5081 |
88
- | 0.4889 | 2.63 | 74 | 0.4974 |
89
- | 0.486 | 2.7 | 76 | 0.5003 |
90
- | 0.4747 | 2.77 | 78 | 0.4917 |
91
- | 0.5028 | 2.84 | 80 | 0.4849 |
92
- | 0.5366 | 2.91 | 82 | 0.4931 |
93
- | 0.5083 | 2.98 | 84 | 0.4986 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
94
 
95
 
96
  ### Framework versions
97
 
98
- - Transformers 4.38.2
99
- - Pytorch 2.2.1+cu121
100
  - Datasets 2.18.0
101
  - Tokenizers 0.15.2
 
15
 
16
  This model is a fine-tuned version of [WKLI22/detr-resnet-50_finetuned_cppe5](https://huggingface.co/WKLI22/detr-resnet-50_finetuned_cppe5) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.3434
19
 
20
  ## Model description
21
 
 
35
 
36
  The following hyperparameters were used during training:
37
  - learning_rate: 1e-05
38
+ - train_batch_size: 20
39
+ - eval_batch_size: 20
40
  - seed: 42
41
+ - gradient_accumulation_steps: 8
42
+ - total_train_batch_size: 160
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
+ - num_epochs: 10
46
  - mixed_precision_training: Native AMP
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
+ | 0.4846 | 0.16 | 10 | 0.5442 |
53
+ | 0.4586 | 0.32 | 20 | 0.4547 |
54
+ | 0.4353 | 0.48 | 30 | 0.4394 |
55
+ | 0.4028 | 0.63 | 40 | 0.4095 |
56
+ | 0.3903 | 0.79 | 50 | 0.4123 |
57
+ | 0.3914 | 0.95 | 60 | 0.4078 |
58
+ | 0.3802 | 1.11 | 70 | 0.3960 |
59
+ | 0.3903 | 1.27 | 80 | 0.3850 |
60
+ | 0.4412 | 1.43 | 90 | 0.3924 |
61
+ | 0.3671 | 1.59 | 100 | 0.3832 |
62
+ | 0.412 | 1.75 | 110 | 0.3629 |
63
+ | 0.3625 | 1.9 | 120 | 0.3593 |
64
+ | 0.3588 | 2.06 | 130 | 0.3533 |
65
+ | 0.3598 | 2.22 | 140 | 0.3525 |
66
+ | 0.404 | 2.38 | 150 | 0.3428 |
67
+ | 0.3519 | 2.54 | 160 | 0.3375 |
68
+ | 0.3647 | 2.7 | 170 | 0.3352 |
69
+ | 0.3669 | 2.86 | 180 | 0.3509 |
70
+ | 0.3695 | 3.02 | 190 | 0.3452 |
71
+ | 0.341 | 3.17 | 200 | 0.3614 |
72
+ | 0.3798 | 3.33 | 210 | 0.3589 |
73
+ | 0.3421 | 3.49 | 220 | 0.3646 |
74
+ | 0.3541 | 3.65 | 230 | 0.3562 |
75
+ | 0.4168 | 3.81 | 240 | 0.3584 |
76
+ | 0.3423 | 3.97 | 250 | 0.3508 |
77
+ | 0.3548 | 4.13 | 260 | 0.3339 |
78
+ | 0.3854 | 4.29 | 270 | 0.3424 |
79
+ | 0.3435 | 4.44 | 280 | 0.3353 |
80
+ | 0.4037 | 4.6 | 290 | 0.3408 |
81
+ | 0.3741 | 4.76 | 300 | 0.3317 |
82
+ | 0.3454 | 4.92 | 310 | 0.3112 |
83
+ | 0.3717 | 5.08 | 320 | 0.3211 |
84
+ | 0.3695 | 5.24 | 330 | 0.3424 |
85
+ | 0.3379 | 5.4 | 340 | 0.3321 |
86
+ | 0.3516 | 5.56 | 350 | 0.3441 |
87
+ | 0.3672 | 5.71 | 360 | 0.3307 |
88
+ | 0.3842 | 5.87 | 370 | 0.3414 |
89
+ | 0.3385 | 6.03 | 380 | 0.3386 |
90
+ | 0.3613 | 6.19 | 390 | 0.3248 |
91
+ | 0.3542 | 6.35 | 400 | 0.3217 |
92
+ | 0.3509 | 6.51 | 410 | 0.3180 |
93
+ | 0.3532 | 6.67 | 420 | 0.3217 |
94
+ | 0.3426 | 6.83 | 430 | 0.3393 |
95
+ | 0.3476 | 6.98 | 440 | 0.3400 |
96
+ | 0.3384 | 7.14 | 450 | 0.3334 |
97
+ | 0.3568 | 7.3 | 460 | 0.3300 |
98
+ | 0.3253 | 7.46 | 470 | 0.3414 |
99
+ | 0.3453 | 7.62 | 480 | 0.3367 |
100
+ | 0.3507 | 7.78 | 490 | 0.3340 |
101
+ | 0.3198 | 7.94 | 500 | 0.3213 |
102
+ | 0.3121 | 8.1 | 510 | 0.3448 |
103
+ | 0.3492 | 8.25 | 520 | 0.3426 |
104
+ | 0.3382 | 8.41 | 530 | 0.3392 |
105
+ | 0.3498 | 8.57 | 540 | 0.3433 |
106
+ | 0.3504 | 8.73 | 550 | 0.3520 |
107
+ | 0.3255 | 8.89 | 560 | 0.3370 |
108
+ | 0.3294 | 9.05 | 570 | 0.3390 |
109
+ | 0.3325 | 9.21 | 580 | 0.3392 |
110
+ | 0.3304 | 9.37 | 590 | 0.3358 |
111
+ | 0.3393 | 9.52 | 600 | 0.3415 |
112
+ | 0.3198 | 9.68 | 610 | 0.3388 |
113
+ | 0.3576 | 9.84 | 620 | 0.3352 |
114
+ | 0.3801 | 10.0 | 630 | 0.3434 |
115
 
116
 
117
  ### Framework versions
118
 
119
+ - Transformers 4.39.3
120
+ - Pytorch 2.2.2+cu121
121
  - Datasets 2.18.0
122
  - Tokenizers 0.15.2
config.json CHANGED
@@ -47,7 +47,7 @@
47
  "position_embedding_type": "sine",
48
  "scale_embedding": false,
49
  "torch_dtype": "float32",
50
- "transformers_version": "4.38.2",
51
  "use_pretrained_backbone": true,
52
  "use_timm_backbone": true
53
  }
 
47
  "position_embedding_type": "sine",
48
  "scale_embedding": false,
49
  "torch_dtype": "float32",
50
+ "transformers_version": "4.39.3",
51
  "use_pretrained_backbone": true,
52
  "use_timm_backbone": true
53
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e6c887eeb192d1471dc373a582b282640c4cd0c2ead96dd6e1c44e9fe63e9331
3
  size 166494824
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bb901a438a1ff82f3f67260519f3be47b3d42bf586432860bbe6af59cac232c
3
  size 166494824
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c4a10acbaab52f350a8e4085ab48ae1d3174b89779ca7ebc3192553df612a4fa
3
  size 4920
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af6e70331c668d3add0c125ed07e353e45663ac0747389cbbd5eb5376bceed23
3
  size 4920