Hugging Face's logo

mrdbourke
/
rt_detrv2_finetuned_trashify_box_detector_v1

Object Detection
Transformers
Safetensors
rt_detr_v2
Generated from Trainer
Model card Files
xet
Community
mrdbourke commited on
Commit
baf78e0
·
verified ·
1 Parent(s): eb832ea

upload fine-tuned RT-DETRv2 trashify object detection model

Browse files
Files changed (4) hide show
  1. README.md +38 -31
  2. config.json +1 -1
  3. model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -16,26 +16,33 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [PekingU/rtdetr_v2_r50vd](https://huggingface.co/PekingU/rtdetr_v2_r50vd) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 9.2205
20
- - Map: 0.4839
21
- - Map 50: 0.6387
22
- - Map 75: 0.5491
23
- - Map Small: 0.0615
24
- - Map Medium: 0.3311
25
- - Map Large: 0.5103
26
- - Mar 1: 0.5719
27
- - Mar 10: 0.7431
28
- - Mar 100: 0.7934
29
  - Mar Small: 0.4
30
- - Mar Medium: 0.629
31
- - Mar Large: 0.8203
32
- - Map Bin: 0.7979
33
- - Map Hand: 0.5783
34
- - Map Not Bin: 0.107
 
 
 
35
  - Map Not Hand: -1.0
36
- - Map Not Trash: 0.2104
37
- - Map Trash: 0.6756
38
- - Map Trash Arm: 0.5341
 
 
 
 
39
 
40
  ## Model description
41
 
@@ -66,23 +73,23 @@ The following hyperparameters were used during training:
66
 
67
  ### Training results
68
 
69
- | Training Loss | Epoch | Step | Validation Loss | Map | Map 50 | Map 75 | Map Small | Map Medium | Map Large | Mar 1 | Mar 10 | Mar 100 | Mar Small | Mar Medium | Mar Large | Map Bin | Map Hand | Map Not Bin | Map Not Hand | Map Not Trash | Map Trash | Map Trash Arm |
70
- |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:----------:|:---------:|:------:|:------:|:-------:|:---------:|:----------:|:---------:|:-------:|:--------:|:-----------:|:------------:|:-------------:|:---------:|:-------------:|
71
- | 40.5747 | 1.0 | 50 | 16.2981 | 0.246 | 0.3366 | 0.2675 | 0.025 | 0.0171 | 0.2549 | 0.3117 | 0.4962 | 0.5861 | 0.05 | 0.2642 | 0.6282 | 0.6576 | 0.4909 | 0.0065 | -1.0 | 0.0364 | 0.2845 | 0.0 |
72
- | 21.4016 | 2.0 | 100 | 11.0868 | 0.3587 | 0.4836 | 0.3921 | 0.0708 | 0.2029 | 0.3824 | 0.467 | 0.6757 | 0.7477 | 0.25 | 0.5415 | 0.7744 | 0.7292 | 0.5612 | 0.0972 | -1.0 | 0.1708 | 0.583 | 0.0108 |
73
- | 17.0522 | 3.0 | 150 | 10.0713 | 0.4282 | 0.5886 | 0.4953 | 0.07 | 0.1468 | 0.4525 | 0.4954 | 0.6945 | 0.7624 | 0.35 | 0.5369 | 0.8008 | 0.7786 | 0.556 | 0.1054 | -1.0 | 0.1919 | 0.6362 | 0.3013 |
74
- | 15.108 | 4.0 | 200 | 9.7776 | 0.417 | 0.5701 | 0.4781 | 0.225 | 0.2119 | 0.4418 | 0.4702 | 0.7301 | 0.7874 | 0.45 | 0.4972 | 0.8222 | 0.7493 | 0.5743 | 0.114 | -1.0 | 0.2339 | 0.6221 | 0.2085 |
75
- | 14.0142 | 5.0 | 250 | 9.4243 | 0.4441 | 0.5981 | 0.5051 | 0.15 | 0.2906 | 0.4747 | 0.5329 | 0.7206 | 0.774 | 0.45 | 0.5631 | 0.8018 | 0.777 | 0.5497 | 0.1943 | -1.0 | 0.2231 | 0.6459 | 0.2743 |
76
- | 12.9659 | 6.0 | 300 | 9.3518 | 0.4646 | 0.6249 | 0.5254 | 0.1357 | 0.3498 | 0.4908 | 0.5683 | 0.7271 | 0.7685 | 0.4 | 0.6085 | 0.7968 | 0.7892 | 0.5766 | 0.1655 | -1.0 | 0.2159 | 0.6596 | 0.381 |
77
- | 12.2048 | 7.0 | 350 | 9.3528 | 0.4886 | 0.6577 | 0.5592 | 0.1333 | 0.2879 | 0.5156 | 0.5362 | 0.719 | 0.779 | 0.4 | 0.5994 | 0.8062 | 0.7776 | 0.5952 | 0.1124 | -1.0 | 0.2089 | 0.6624 | 0.5753 |
78
- | 11.5839 | 8.0 | 400 | 9.3688 | 0.4524 | 0.615 | 0.5207 | 0.1208 | 0.2991 | 0.4775 | 0.5596 | 0.7403 | 0.7853 | 0.4 | 0.6062 | 0.8157 | 0.7882 | 0.5613 | 0.1261 | -1.0 | 0.1989 | 0.6574 | 0.3827 |
79
- | 10.9767 | 9.0 | 450 | 9.3092 | 0.4834 | 0.6392 | 0.5463 | 0.0667 | 0.309 | 0.512 | 0.5599 | 0.7268 | 0.7886 | 0.4 | 0.6034 | 0.8159 | 0.7888 | 0.5736 | 0.1552 | -1.0 | 0.205 | 0.6702 | 0.5076 |
80
- | 10.609 | 10.0 | 500 | 9.2205 | 0.4839 | 0.6387 | 0.5491 | 0.0615 | 0.3311 | 0.5103 | 0.5719 | 0.7431 | 0.7934 | 0.4 | 0.629 | 0.8203 | 0.7979 | 0.5783 | 0.107 | -1.0 | 0.2104 | 0.6756 | 0.5341 |
81
 
82
 
83
  ### Framework versions
84
 
85
- - Transformers 4.52.0.dev0
86
  - Pytorch 2.7.0+cu126
87
  - Datasets 3.6.0
88
  - Tokenizers 0.21.1
 
16
 
17
  This model is a fine-tuned version of [PekingU/rtdetr_v2_r50vd](https://huggingface.co/PekingU/rtdetr_v2_r50vd) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 8.9000
20
+ - Map: 0.5134
21
+ - Map 50: 0.6917
22
+ - Map 75: 0.5749
23
+ - Map Small: 0.4
24
+ - Map Medium: 0.2845
25
+ - Map Large: 0.5538
26
+ - Mar 1: 0.5482
27
+ - Mar 10: 0.7189
28
+ - Mar 100: 0.7663
29
  - Mar Small: 0.4
30
+ - Mar Medium: 0.533
31
+ - Mar Large: 0.7931
32
+ - Map Bin: 0.7876
33
+ - Mar 100 Bin: 0.8879
34
+ - Map Hand: 0.5723
35
+ - Mar 100 Hand: 0.8118
36
+ - Map Not Bin: 0.1797
37
+ - Mar 100 Not Bin: 0.6857
38
  - Map Not Hand: -1.0
39
+ - Mar 100 Not Hand: -1.0
40
+ - Map Not Trash: 0.2679
41
+ - Mar 100 Not Trash: 0.625
42
+ - Map Trash: 0.6726
43
+ - Mar 100 Trash: 0.7876
44
+ - Map Trash Arm: 0.6
45
+ - Mar 100 Trash Arm: 0.8
46
 
47
  ## Model description
48
 
 
73
 
74
  ### Training results
75
 
76
+ | Training Loss | Epoch | Step | Validation Loss | Map | Map 50 | Map 75 | Map Small | Map Medium | Map Large | Mar 1 | Mar 10 | Mar 100 | Mar Small | Mar Medium | Mar Large | Map Bin | Mar 100 Bin | Map Hand | Mar 100 Hand | Map Not Bin | Mar 100 Not Bin | Map Not Hand | Mar 100 Not Hand | Map Not Trash | Mar 100 Not Trash | Map Trash | Mar 100 Trash | Map Trash Arm | Mar 100 Trash Arm |
77
+ |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:----------:|:---------:|:------:|:------:|:-------:|:---------:|:----------:|:---------:|:-------:|:-----------:|:--------:|:------------:|:-----------:|:---------------:|:------------:|:----------------:|:-------------:|:-----------------:|:---------:|:-------------:|:-------------:|:-----------------:|
78
+ | 75.2499 | 1.0 | 50 | 17.5113 | 0.2036 | 0.293 | 0.2137 | 0.0 | 0.0349 | 0.2153 | 0.2926 | 0.4248 | 0.508 | 0.0 | 0.1244 | 0.5579 | 0.5792 | 0.8312 | 0.2434 | 0.7696 | 0.0044 | 0.3429 | -1.0 | -1.0 | 0.0107 | 0.4639 | 0.3837 | 0.6407 | 0.0 | 0.0 |
79
+ | 23.852 | 2.0 | 100 | 11.4502 | 0.2711 | 0.3799 | 0.3015 | 0.05 | 0.1059 | 0.2818 | 0.3735 | 0.5918 | 0.6483 | 0.35 | 0.3608 | 0.6945 | 0.6972 | 0.9035 | 0.2595 | 0.8088 | 0.0109 | 0.5643 | -1.0 | -1.0 | 0.031 | 0.5958 | 0.6088 | 0.7504 | 0.0192 | 0.2667 |
80
+ | 18.2873 | 3.0 | 150 | 10.0729 | 0.4112 | 0.5678 | 0.4869 | 0.3655 | 0.2303 | 0.432 | 0.4785 | 0.6951 | 0.7657 | 0.45 | 0.4551 | 0.7968 | 0.7569 | 0.905 | 0.3534 | 0.8343 | 0.0278 | 0.6571 | -1.0 | -1.0 | 0.1497 | 0.6236 | 0.6421 | 0.7743 | 0.5371 | 0.8 |
81
+ | 15.8982 | 4.0 | 200 | 9.4929 | 0.48 | 0.6555 | 0.5578 | 0.4 | 0.2552 | 0.5051 | 0.524 | 0.7099 | 0.7588 | 0.4 | 0.4597 | 0.7931 | 0.753 | 0.8936 | 0.5989 | 0.8353 | 0.1333 | 0.6429 | -1.0 | -1.0 | 0.1993 | 0.6319 | 0.6537 | 0.7823 | 0.542 | 0.7667 |
82
+ | 14.6758 | 5.0 | 250 | 9.4786 | 0.47 | 0.6472 | 0.5411 | 0.4 | 0.2494 | 0.5009 | 0.5346 | 0.6907 | 0.7252 | 0.4 | 0.3784 | 0.7732 | 0.7641 | 0.8766 | 0.5657 | 0.8029 | 0.1636 | 0.5571 | -1.0 | -1.0 | 0.2588 | 0.6083 | 0.6364 | 0.7726 | 0.4312 | 0.7333 |
83
+ | 13.5443 | 6.0 | 300 | 9.2135 | 0.495 | 0.6699 | 0.5594 | 0.35 | 0.347 | 0.5225 | 0.5432 | 0.7086 | 0.7602 | 0.35 | 0.5625 | 0.7905 | 0.7808 | 0.895 | 0.5788 | 0.8157 | 0.1336 | 0.6286 | -1.0 | -1.0 | 0.2336 | 0.6208 | 0.6626 | 0.8009 | 0.5804 | 0.8 |
84
+ | 12.828 | 7.0 | 350 | 8.9653 | 0.5041 | 0.6851 | 0.5799 | 0.35 | 0.2242 | 0.5328 | 0.543 | 0.7152 | 0.7596 | 0.35 | 0.5034 | 0.7952 | 0.7919 | 0.8922 | 0.5883 | 0.8127 | 0.1407 | 0.6643 | -1.0 | -1.0 | 0.2459 | 0.6264 | 0.6884 | 0.7956 | 0.5692 | 0.7667 |
85
+ | 12.1564 | 8.0 | 400 | 8.8797 | 0.509 | 0.683 | 0.5708 | 0.35 | 0.2002 | 0.542 | 0.5565 | 0.7412 | 0.7722 | 0.35 | 0.5267 | 0.8006 | 0.782 | 0.8879 | 0.6009 | 0.8137 | 0.1517 | 0.6857 | -1.0 | -1.0 | 0.2626 | 0.6278 | 0.6564 | 0.785 | 0.6003 | 0.8333 |
86
+ | 11.5731 | 9.0 | 450 | 9.0043 | 0.5126 | 0.692 | 0.5879 | 0.4 | 0.2861 | 0.5548 | 0.5454 | 0.7211 | 0.7714 | 0.4 | 0.5199 | 0.8015 | 0.7828 | 0.8823 | 0.5674 | 0.8176 | 0.2052 | 0.6929 | -1.0 | -1.0 | 0.2661 | 0.6139 | 0.6843 | 0.7885 | 0.5698 | 0.8333 |
87
+ | 11.2251 | 10.0 | 500 | 8.9000 | 0.5134 | 0.6917 | 0.5749 | 0.4 | 0.2845 | 0.5538 | 0.5482 | 0.7189 | 0.7663 | 0.4 | 0.533 | 0.7931 | 0.7876 | 0.8879 | 0.5723 | 0.8118 | 0.1797 | 0.6857 | -1.0 | -1.0 | 0.2679 | 0.625 | 0.6726 | 0.7876 | 0.6 | 0.8 |
88
 
89
 
90
  ### Framework versions
91
 
92
+ - Transformers 4.52.3
93
  - Pytorch 2.7.0+cu126
94
  - Datasets 3.6.0
95
  - Tokenizers 0.21.1
config.json CHANGED
@@ -125,7 +125,7 @@
125
  "num_queries": 300,
126
  "positional_encoding_temperature": 10000,
127
  "torch_dtype": "float32",
128
- "transformers_version": "4.52.0.dev0",
129
  "use_focal_loss": true,
130
  "use_pretrained_backbone": false,
131
  "use_timm_backbone": false,
 
125
  "num_queries": 300,
126
  "positional_encoding_temperature": 10000,
127
  "torch_dtype": "float32",
128
+ "transformers_version": "4.52.3",
129
  "use_focal_loss": true,
130
  "use_pretrained_backbone": false,
131
  "use_timm_backbone": false,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:810f0b7727fccef056e211ca8fd7f4db42d3829a9108ced41f6709dab0c36d0d
3
  size 171576780
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7419f7006fb3d491453145609d63546a304c17354697d137408d23242ffabe52
3
  size 171576780
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:402df5ed3b398b4e71df6e56fa61214202196f07f86ba8806505c5cc3fc9a386
3
  size 5777
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3782ebdc15a8999842a98f0e26e37cb1bb7b65e6e2adba862d5d55f94a6079c5
3
  size 5777