Hugging Face's logo

SARANGx
/
rtdetr-pill-detector

Object Detection
Transformers
Safetensors
rt_detr
pill-detection
medical
rt-detr
computer-vision
counting
healthcare
Eval Results (legacy)
Model card Files
xet
 Community
SARANGx commited on
Commit
f019ecd
·
verified ·
1 Parent(s): 6ab02fb

Training complete - RT-DETR R18 pill detector, 20 epochs

Browse files
Files changed (3) hide show
  1. README.md +24 -24
  2. model.safetensors +1 -1
  3. training_args.bin +2 -2
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 [PekingU/rtdetr_r18vd_coco_o365](https://huggingface.co/PekingU/rtdetr_r18vd_coco_o365) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 2.5969
20
 
21
  ## Model description
22
 
@@ -36,38 +36,38 @@ More information needed
36
 
37
  The following hyperparameters were used during training:
38
  - learning_rate: 5e-05
39
- - train_batch_size: 4
40
- - eval_batch_size: 4
41
  - seed: 42
42
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
43
  - lr_scheduler_type: cosine
44
- - lr_scheduler_warmup_steps: 100
45
  - num_epochs: 20
46
 
47
  ### Training results
48
 
49
  | Training Loss | Epoch | Step | Validation Loss |
50
  |:-------------:|:-----:|:----:|:---------------:|
51
- | 49.1683 | 1.0 | 79 | 34.2370 |
52
- | 18.6682 | 2.0 | 158 | 11.6126 |
53
- | 12.7908 | 3.0 | 237 | 6.9794 |
54
- | 10.8112 | 4.0 | 316 | 5.3765 |
55
- | 8.4986 | 5.0 | 395 | 4.3265 |
56
- | 7.0638 | 6.0 | 474 | 3.7277 |
57
- | 6.9021 | 7.0 | 553 | 3.4731 |
58
- | 6.5311 | 8.0 | 632 | 2.9624 |
59
- | 6.5105 | 9.0 | 711 | 3.0513 |
60
- | 6.2654 | 10.0 | 790 | 2.7852 |
61
- | 5.9322 | 11.0 | 869 | 2.9419 |
62
- | 5.2394 | 12.0 | 948 | 2.8256 |
63
- | 5.7140 | 13.0 | 1027 | 2.7302 |
64
- | 6.3316 | 14.0 | 1106 | 2.6021 |
65
- | 5.0378 | 15.0 | 1185 | 2.6694 |
66
- | 5.2466 | 16.0 | 1264 | 2.6375 |
67
- | 5.3730 | 17.0 | 1343 | 2.6533 |
68
- | 4.9664 | 18.0 | 1422 | 2.6242 |
69
- | 5.2872 | 19.0 | 1501 | 2.6111 |
70
- | 5.5504 | 20.0 | 1580 | 2.5969 |
71
 
72
 
73
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [PekingU/rtdetr_r18vd_coco_o365](https://huggingface.co/PekingU/rtdetr_r18vd_coco_o365) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 3.5814
20
 
21
  ## Model description
22
 
 
36
 
37
  The following hyperparameters were used during training:
38
  - learning_rate: 5e-05
39
+ - train_batch_size: 8
40
+ - eval_batch_size: 8
41
  - seed: 42
42
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
43
  - lr_scheduler_type: cosine
44
+ - lr_scheduler_warmup_steps: 50
45
  - num_epochs: 20
46
 
47
  ### Training results
48
 
49
  | Training Loss | Epoch | Step | Validation Loss |
50
  |:-------------:|:-----:|:----:|:---------------:|
51
+ | 36.6133 | 1.0 | 40 | 25.6964 |
52
+ | 16.9363 | 2.0 | 80 | 10.3644 |
53
+ | 13.2576 | 3.0 | 120 | 7.4604 |
54
+ | 11.2193 | 4.0 | 160 | 6.0646 |
55
+ | 8.5718 | 5.0 | 200 | 5.2125 |
56
+ | 8.3696 | 6.0 | 240 | 4.8302 |
57
+ | 7.5242 | 7.0 | 280 | 4.4670 |
58
+ | 7.0528 | 8.0 | 320 | 4.0657 |
59
+ | 6.3887 | 9.0 | 360 | 3.8813 |
60
+ | 6.3234 | 10.0 | 400 | 3.7910 |
61
+ | 5.9830 | 11.0 | 440 | 3.6750 |
62
+ | 5.8632 | 12.0 | 480 | 3.5891 |
63
+ | 6.7105 | 13.0 | 520 | 3.7327 |
64
+ | 5.4882 | 14.0 | 560 | 3.5292 |
65
+ | 5.5451 | 15.0 | 600 | 3.5306 |
66
+ | 5.3121 | 16.0 | 640 | 3.5427 |
67
+ | 5.2448 | 17.0 | 680 | 3.5641 |
68
+ | 5.3653 | 18.0 | 720 | 3.6307 |
69
+ | 5.3143 | 19.0 | 760 | 3.5280 |
70
+ | 5.7803 | 20.0 | 800 | 3.5814 |
71
 
72
 
73
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:14f14b9adead20e2210f469516f72a6d6c6bae1c64b8c955b1bf2ad880f30bfc
3
  size 80539488
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04066119b4e46eaae24d574aef40792b4d242bc4ed8904c5831a006cfeb80515
3
  size 80539488
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0b4170760098cff9fafecd65cd7d93f4f8aec7d55ea50c4d43ee5b9443eb8cd1
3
- size 5265
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2df32deb97154fb09b69fe5229692409299f9b7946b277c08936b8c62ba936a0
3
+ size 5329