Hugging Face's logo

mrdbourke
/
detr_finetuned_trashify_box_detector

Object Detection
Transformers
Safetensors
conditional_detr
Generated from Trainer
Inference Endpoints
Model card Files Community
mrdbourke commited on
Commit
3e55c42
·
verified ·
1 Parent(s): 564dbf4

upload trashify object detection model

Browse files
Files changed (5) hide show
  1. README.md +31 -105
  2. config.json +5 -3
  3. model.safetensors +2 -2
  4. preprocessor_config.json +3 -6
  5. training_args.bin +2 -2
README.md CHANGED
@@ -1,4 +1,5 @@
1
  ---
 
2
  license: apache-2.0
3
  base_model: microsoft/conditional-detr-resnet-50
4
  tags:
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  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.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 1.2291
19
 
20
  ## Model description
21
 
@@ -41,118 +42,43 @@ The following hyperparameters were used during training:
41
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
  - lr_scheduler_type: linear
43
  - lr_scheduler_warmup_ratio: 0.05
44
- - num_epochs: 100
45
  - mixed_precision_training: Native AMP
46
 
47
  ### Training results
48
 
49
  | Training Loss | Epoch | Step | Validation Loss |
50
  |:-------------:|:-----:|:----:|:---------------:|
51
- | 137.9231 | 1.0 | 17 | 125.8028 |
52
- | 84.2349 | 2.0 | 34 | 40.2665 |
53
- | 19.7645 | 3.0 | 51 | 7.3174 |
54
- | 5.1905 | 4.0 | 68 | 3.5275 |
55
- | 3.2078 | 5.0 | 85 | 2.6258 |
56
- | 2.5905 | 6.0 | 102 | 2.2457 |
57
- | 2.3692 | 7.0 | 119 | 2.3092 |
58
- | 2.2272 | 8.0 | 136 | 1.9838 |
59
- | 2.1079 | 9.0 | 153 | 1.8935 |
60
- | 2.0051 | 10.0 | 170 | 1.7939 |
61
- | 1.8845 | 11.0 | 187 | 1.7864 |
62
- | 1.8544 | 12.0 | 204 | 1.6802 |
63
- | 1.7695 | 13.0 | 221 | 1.5136 |
64
- | 1.6797 | 14.0 | 238 | 1.5392 |
65
- | 1.5638 | 15.0 | 255 | 1.4511 |
66
- | 1.5424 | 16.0 | 272 | 1.4155 |
67
- | 1.4715 | 17.0 | 289 | 1.4013 |
68
- | 1.4396 | 18.0 | 306 | 1.3514 |
69
- | 1.4471 | 19.0 | 323 | 1.3247 |
70
- | 1.3963 | 20.0 | 340 | 1.2975 |
71
- | 1.3261 | 21.0 | 357 | 1.2779 |
72
- | 1.2633 | 22.0 | 374 | 1.3002 |
73
- | 1.2494 | 23.0 | 391 | 1.2980 |
74
- | 1.2512 | 24.0 | 408 | 1.2877 |
75
- | 1.2433 | 25.0 | 425 | 1.2777 |
76
- | 1.2116 | 26.0 | 442 | 1.2703 |
77
- | 1.2489 | 27.0 | 459 | 1.2643 |
78
- | 1.2316 | 28.0 | 476 | 1.2370 |
79
- | 1.1575 | 29.0 | 493 | 1.2619 |
80
- | 1.1525 | 30.0 | 510 | 1.1912 |
81
- | 1.1134 | 31.0 | 527 | 1.2454 |
82
- | 1.1722 | 32.0 | 544 | 1.2335 |
83
- | 1.1395 | 33.0 | 561 | 1.2095 |
84
- | 1.1626 | 34.0 | 578 | 1.2022 |
85
- | 1.0886 | 35.0 | 595 | 1.1997 |
86
- | 1.0653 | 36.0 | 612 | 1.2023 |
87
- | 1.0653 | 37.0 | 629 | 1.1955 |
88
- | 1.0421 | 38.0 | 646 | 1.2127 |
89
- | 1.0476 | 39.0 | 663 | 1.1780 |
90
- | 1.0415 | 40.0 | 680 | 1.2002 |
91
- | 1.0107 | 41.0 | 697 | 1.1691 |
92
- | 0.9861 | 42.0 | 714 | 1.2002 |
93
- | 1.0084 | 43.0 | 731 | 1.1759 |
94
- | 1.0171 | 44.0 | 748 | 1.1726 |
95
- | 0.9977 | 45.0 | 765 | 1.1668 |
96
- | 0.9553 | 46.0 | 782 | 1.2050 |
97
- | 0.9872 | 47.0 | 799 | 1.1835 |
98
- | 0.9529 | 48.0 | 816 | 1.2004 |
99
- | 0.9669 | 49.0 | 833 | 1.1595 |
100
- | 0.9669 | 50.0 | 850 | 1.1474 |
101
- | 0.9218 | 51.0 | 867 | 1.1900 |
102
- | 0.9517 | 52.0 | 884 | 1.1845 |
103
- | 0.9665 | 53.0 | 901 | 1.1831 |
104
- | 0.8924 | 54.0 | 918 | 1.2106 |
105
- | 0.9008 | 55.0 | 935 | 1.1995 |
106
- | 0.8584 | 56.0 | 952 | 1.2195 |
107
- | 0.9453 | 57.0 | 969 | 1.2342 |
108
- | 0.915 | 58.0 | 986 | 1.2210 |
109
- | 0.9046 | 59.0 | 1003 | 1.2073 |
110
- | 0.8851 | 60.0 | 1020 | 1.1895 |
111
- | 0.9139 | 61.0 | 1037 | 1.2091 |
112
- | 0.8388 | 62.0 | 1054 | 1.2165 |
113
- | 0.8799 | 63.0 | 1071 | 1.2143 |
114
- | 0.8534 | 64.0 | 1088 | 1.2154 |
115
- | 0.8412 | 65.0 | 1105 | 1.2134 |
116
- | 0.8305 | 66.0 | 1122 | 1.2277 |
117
- | 0.8485 | 67.0 | 1139 | 1.2750 |
118
- | 0.8177 | 68.0 | 1156 | 1.2791 |
119
- | 0.8266 | 69.0 | 1173 | 1.2348 |
120
- | 0.8448 | 70.0 | 1190 | 1.2564 |
121
- | 0.8392 | 71.0 | 1207 | 1.2211 |
122
- | 0.8007 | 72.0 | 1224 | 1.2272 |
123
- | 0.8609 | 73.0 | 1241 | 1.2467 |
124
- | 0.838 | 74.0 | 1258 | 1.2389 |
125
- | 0.8095 | 75.0 | 1275 | 1.2609 |
126
- | 0.7726 | 76.0 | 1292 | 1.2399 |
127
- | 0.7956 | 77.0 | 1309 | 1.2718 |
128
- | 0.8253 | 78.0 | 1326 | 1.2352 |
129
- | 0.7387 | 79.0 | 1343 | 1.2423 |
130
- | 0.8324 | 80.0 | 1360 | 1.2692 |
131
- | 0.77 | 81.0 | 1377 | 1.2384 |
132
- | 0.8133 | 82.0 | 1394 | 1.2354 |
133
- | 0.7822 | 83.0 | 1411 | 1.2231 |
134
- | 0.7512 | 84.0 | 1428 | 1.2178 |
135
- | 0.7264 | 85.0 | 1445 | 1.2305 |
136
- | 0.7814 | 86.0 | 1462 | 1.2202 |
137
- | 0.7954 | 87.0 | 1479 | 1.2102 |
138
- | 0.7679 | 88.0 | 1496 | 1.2158 |
139
- | 0.7577 | 89.0 | 1513 | 1.2429 |
140
- | 0.7122 | 90.0 | 1530 | 1.2462 |
141
- | 0.7859 | 91.0 | 1547 | 1.2340 |
142
- | 0.7276 | 92.0 | 1564 | 1.2447 |
143
- | 0.7651 | 93.0 | 1581 | 1.2286 |
144
- | 0.743 | 94.0 | 1598 | 1.2402 |
145
- | 0.7813 | 95.0 | 1615 | 1.2276 |
146
- | 0.6978 | 96.0 | 1632 | 1.2364 |
147
- | 0.6845 | 97.0 | 1649 | 1.2392 |
148
- | 0.7037 | 98.0 | 1666 | 1.2367 |
149
- | 0.7468 | 99.0 | 1683 | 1.2323 |
150
- | 0.7241 | 100.0 | 1700 | 1.2291 |
151
 
152
 
153
  ### Framework versions
154
 
155
- - Transformers 4.44.0
156
- - Pytorch 2.2.0+cu121
157
- - Datasets 2.20.0
158
  - Tokenizers 0.19.1
 
1
  ---
2
+ library_name: transformers
3
  license: apache-2.0
4
  base_model: microsoft/conditional-detr-resnet-50
5
  tags:
 
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: 1.0728
20
 
21
  ## Model description
22
 
 
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
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
+ | 2.6374 | 1.0 | 25 | 2.2767 |
53
+ | 2.3806 | 2.0 | 50 | 2.0779 |
54
+ | 2.17 | 3.0 | 75 | 1.8047 |
55
+ | 1.9959 | 4.0 | 100 | 1.7031 |
56
+ | 1.9129 | 5.0 | 125 | 1.6561 |
57
+ | 1.7917 | 6.0 | 150 | 1.5245 |
58
+ | 1.6619 | 7.0 | 175 | 1.4002 |
59
+ | 1.5913 | 8.0 | 200 | 1.3501 |
60
+ | 1.4466 | 9.0 | 225 | 1.2675 |
61
+ | 1.3785 | 10.0 | 250 | 1.2594 |
62
+ | 1.3542 | 11.0 | 275 | 1.2620 |
63
+ | 1.349 | 12.0 | 300 | 1.1957 |
64
+ | 1.3055 | 13.0 | 325 | 1.1818 |
65
+ | 1.1731 | 14.0 | 350 | 1.1466 |
66
+ | 1.1585 | 15.0 | 375 | 1.1627 |
67
+ | 1.0614 | 16.0 | 400 | 1.1615 |
68
+ | 1.0144 | 17.0 | 425 | 1.1411 |
69
+ | 0.967 | 18.0 | 450 | 1.1193 |
70
+ | 0.9114 | 19.0 | 475 | 1.1043 |
71
+ | 0.9031 | 20.0 | 500 | 1.1005 |
72
+ | 0.8719 | 21.0 | 525 | 1.1000 |
73
+ | 0.8505 | 22.0 | 550 | 1.0800 |
74
+ | 0.8314 | 23.0 | 575 | 1.0739 |
75
+ | 0.8178 | 24.0 | 600 | 1.0757 |
76
+ | 0.804 | 25.0 | 625 | 1.0728 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
77
 
78
 
79
  ### Framework versions
80
 
81
+ - Transformers 4.45.0.dev0
82
+ - Pytorch 2.4.0+cu124
83
+ - Datasets 2.21.0
84
  - Tokenizers 0.19.1
config.json CHANGED
@@ -43,7 +43,8 @@
43
  "2": "not_bin",
44
  "3": "not_hand",
45
  "4": "not_trash",
46
- "5": "trash"
 
47
  },
48
  "init_std": 0.02,
49
  "init_xavier_std": 1.0,
@@ -54,7 +55,8 @@
54
  "not_bin": 2,
55
  "not_hand": 3,
56
  "not_trash": 4,
57
- "trash": 5
 
58
  },
59
  "mask_loss_coefficient": 1,
60
  "max_position_embeddings": 1024,
@@ -65,7 +67,7 @@
65
  "position_embedding_type": "sine",
66
  "scale_embedding": false,
67
  "torch_dtype": "float32",
68
- "transformers_version": "4.44.0",
69
  "use_pretrained_backbone": true,
70
  "use_timm_backbone": true
71
  }
 
43
  "2": "not_bin",
44
  "3": "not_hand",
45
  "4": "not_trash",
46
+ "5": "trash",
47
+ "6": "trash_arm"
48
  },
49
  "init_std": 0.02,
50
  "init_xavier_std": 1.0,
 
55
  "not_bin": 2,
56
  "not_hand": 3,
57
  "not_trash": 4,
58
+ "trash": 5,
59
+ "trash_arm": 6
60
  },
61
  "mask_loss_coefficient": 1,
62
  "max_position_embeddings": 1024,
 
67
  "position_embedding_type": "sine",
68
  "scale_embedding": false,
69
  "torch_dtype": "float32",
70
+ "transformers_version": "4.45.0.dev0",
71
  "use_pretrained_backbone": true,
72
  "use_timm_backbone": true
73
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5715f9b8e8a7514bfec7282ae3bcf3c730de0c8e0d2126ddd606870656444d83
3
- size 174080824
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d50b5beb7b5f8a524a56542ed9793be6fa4dbe61cf73ba5eeac0a97e2a218c1a
3
+ size 174081852
preprocessor_config.json CHANGED
@@ -16,14 +16,11 @@
16
  0.224,
17
  0.225
18
  ],
19
- "pad_size": {
20
- "height": 480,
21
- "width": 480
22
- },
23
  "resample": 2,
24
  "rescale_factor": 0.00392156862745098,
25
  "size": {
26
- "height": 480,
27
- "width": 480
28
  }
29
  }
 
16
  0.224,
17
  0.225
18
  ],
19
+ "pad_size": null,
 
 
 
20
  "resample": 2,
21
  "rescale_factor": 0.00392156862745098,
22
  "size": {
23
+ "longest_edge": 640,
24
+ "shortest_edge": 640
25
  }
26
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:74864e64adc55d2bc4d7254004d0c05ebf04ea6b35b11e2f362274faea998721
3
- size 5176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a03ea92e513daa5b2ede961727b411077de93967808efeb6d3214da08fb083be
3
+ size 5240