Hugging Face's logo

mrdbourke
/
detr_finetuned_trashify_box_detector_with_data_aug

Object Detection
Transformers
Safetensors
conditional_detr
Generated from Trainer
Inference Endpoints
Model card Files Community
mrdbourke commited on
Commit
2f5f3ed
•
1 Parent(s): e9790ee

upload trashify object detection model with data augmentation

Browse files
Files changed (3) hide show
  1. README.md +27 -27
  2. model.safetensors +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [microsoft/conditional-detr-resnet-50](https://huggingface.co/microsoft/conditional-detr-resnet-50) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 1.0875
20
 
21
  ## Model description
22
 
@@ -40,7 +40,7 @@ The following hyperparameters were used during training:
40
  - eval_batch_size: 16
41
  - seed: 42
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
- - lr_scheduler_type: cosine
44
  - lr_scheduler_warmup_ratio: 0.05
45
  - num_epochs: 25
46
  - mixed_precision_training: Native AMP
@@ -49,31 +49,31 @@ The following hyperparameters were used during training:
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
- | 80.0272 | 1.0 | 50 | 8.2323 |
53
- | 4.2154 | 2.0 | 100 | 2.7596 |
54
- | 2.5841 | 3.0 | 150 | 2.2609 |
55
- | 2.2857 | 4.0 | 200 | 2.1303 |
56
- | 2.1495 | 5.0 | 250 | 1.8199 |
57
- | 1.9145 | 6.0 | 300 | 1.6257 |
58
- | 1.8117 | 7.0 | 350 | 1.4758 |
59
- | 1.6692 | 8.0 | 400 | 1.4397 |
60
- | 1.6048 | 9.0 | 450 | 1.3883 |
61
- | 1.517 | 10.0 | 500 | 1.3472 |
62
- | 1.4842 | 11.0 | 550 | 1.2953 |
63
- | 1.3999 | 12.0 | 600 | 1.2637 |
64
- | 1.3551 | 13.0 | 650 | 1.2034 |
65
- | 1.3283 | 14.0 | 700 | 1.1875 |
66
- | 1.2783 | 15.0 | 750 | 1.1679 |
67
- | 1.2577 | 16.0 | 800 | 1.1578 |
68
- | 1.1815 | 17.0 | 850 | 1.1271 |
69
- | 1.1911 | 18.0 | 900 | 1.1192 |
70
- | 1.1307 | 19.0 | 950 | 1.1172 |
71
- | 1.1604 | 20.0 | 1000 | 1.1065 |
72
- | 1.1365 | 21.0 | 1050 | 1.1013 |
73
- | 1.1287 | 22.0 | 1100 | 1.0867 |
74
- | 1.1201 | 23.0 | 1150 | 1.0822 |
75
- | 1.1075 | 24.0 | 1200 | 1.0890 |
76
- | 1.1138 | 25.0 | 1250 | 1.0875 |
77
 
78
 
79
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [microsoft/conditional-detr-resnet-50](https://huggingface.co/microsoft/conditional-detr-resnet-50) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 1.0704
20
 
21
  ## Model description
22
 
 
40
  - eval_batch_size: 16
41
  - seed: 42
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
+ - lr_scheduler_type: linear
44
  - lr_scheduler_warmup_ratio: 0.05
45
  - num_epochs: 25
46
  - mixed_precision_training: Native AMP
 
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
+ | 100.4735 | 1.0 | 50 | 8.0297 |
53
+ | 4.369 | 2.0 | 100 | 2.7376 |
54
+ | 2.5518 | 3.0 | 150 | 2.1839 |
55
+ | 2.2226 | 4.0 | 200 | 1.9228 |
56
+ | 1.9906 | 5.0 | 250 | 1.7408 |
57
+ | 1.8219 | 6.0 | 300 | 1.5573 |
58
+ | 1.6974 | 7.0 | 350 | 1.4779 |
59
+ | 1.6027 | 8.0 | 400 | 1.4510 |
60
+ | 1.5517 | 9.0 | 450 | 1.3711 |
61
+ | 1.4491 | 10.0 | 500 | 1.3177 |
62
+ | 1.4335 | 11.0 | 550 | 1.2811 |
63
+ | 1.3645 | 12.0 | 600 | 1.2475 |
64
+ | 1.3314 | 13.0 | 650 | 1.2060 |
65
+ | 1.2973 | 14.0 | 700 | 1.1874 |
66
+ | 1.2506 | 15.0 | 750 | 1.1794 |
67
+ | 1.2319 | 16.0 | 800 | 1.1657 |
68
+ | 1.1479 | 17.0 | 850 | 1.1300 |
69
+ | 1.1466 | 18.0 | 900 | 1.1179 |
70
+ | 1.1138 | 19.0 | 950 | 1.1095 |
71
+ | 1.1153 | 20.0 | 1000 | 1.0961 |
72
+ | 1.0894 | 21.0 | 1050 | 1.0790 |
73
+ | 1.0691 | 22.0 | 1100 | 1.0870 |
74
+ | 1.0619 | 23.0 | 1150 | 1.0804 |
75
+ | 1.0459 | 24.0 | 1200 | 1.0717 |
76
+ | 1.0363 | 25.0 | 1250 | 1.0704 |
77
 
78
 
79
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bedac034656a3787138d606112f3d59df950b00eed884521ebb2e44dc33666a3
3
  size 174081852
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:809090d93d481dfdb71a39b2e875c5db75881eb8ed42c945299a33ae225cae4d
3
  size 174081852
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c107be7e21f9920ec6f0f2c53ee3750417900a8523b0310989b50ce4b09f2e51
3
  size 5240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60ed695b761c770df13cf4e0a2f746953e306d2f29e326067e04a139436efccf
3
  size 5240