detr_finetuned_cppe5
This model is a fine-tuned version of microsoft/conditional-detr-resnet-50 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2426
- Map: 0.268
- Map 50: 0.5294
- Map 75: 0.2419
- Map Small: 0.1163
- Map Medium: 0.2288
- Map Large: 0.5006
- Mar 1: 0.2865
- Mar 10: 0.4475
- Mar 100: 0.4749
- Mar Small: 0.3002
- Mar Medium: 0.4623
- Mar Large: 0.7345
- Map Coverall: 0.5546
- Mar 100 Coverall: 0.6736
- Map Face Shield: 0.1674
- Mar 100 Face Shield: 0.4833
- Map Gloves: 0.1944
- Mar 100 Gloves: 0.3662
- Map Goggles: 0.1199
- Mar 100 Goggles: 0.4421
- Map Mask: 0.3036
- Mar 100 Mask: 0.4092
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 30
Training results
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 Coverall | Mar 100 Coverall | Map Face Shield | Mar 100 Face Shield | Map Gloves | Mar 100 Gloves | Map Goggles | Mar 100 Goggles | Map Mask | Mar 100 Mask |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
No log | 1.0 | 102 | 2.3847 | 0.008 | 0.0259 | 0.0031 | 0.0071 | 0.0113 | 0.0323 | 0.0303 | 0.1094 | 0.1661 | 0.1002 | 0.1711 | 0.2087 | 0.0105 | 0.207 | 0.0034 | 0.0681 | 0.0078 | 0.2167 | 0.0025 | 0.0807 | 0.0156 | 0.2583 |
No log | 2.0 | 204 | 2.1346 | 0.0292 | 0.0819 | 0.0147 | 0.0126 | 0.0268 | 0.042 | 0.0773 | 0.1631 | 0.2064 | 0.0815 | 0.1912 | 0.2543 | 0.0873 | 0.3791 | 0.0083 | 0.1153 | 0.0116 | 0.2069 | 0.0 | 0.0 | 0.0388 | 0.3306 |
No log | 3.0 | 306 | 2.0183 | 0.0546 | 0.1289 | 0.0413 | 0.0288 | 0.0352 | 0.0606 | 0.1234 | 0.2329 | 0.2765 | 0.1078 | 0.2436 | 0.3393 | 0.1671 | 0.5567 | 0.0187 | 0.2167 | 0.0094 | 0.2157 | 0.0346 | 0.0912 | 0.0431 | 0.3024 |
No log | 4.0 | 408 | 1.9305 | 0.0779 | 0.1742 | 0.0587 | 0.02 | 0.0476 | 0.1114 | 0.1417 | 0.2692 | 0.3059 | 0.0926 | 0.2783 | 0.4861 | 0.27 | 0.59 | 0.0335 | 0.3208 | 0.0214 | 0.2377 | 0.0036 | 0.086 | 0.061 | 0.2951 |
3.3072 | 5.0 | 510 | 1.7155 | 0.1186 | 0.2742 | 0.0935 | 0.0303 | 0.0871 | 0.2195 | 0.1434 | 0.3145 | 0.3401 | 0.1573 | 0.3076 | 0.5985 | 0.3762 | 0.5582 | 0.0213 | 0.2569 | 0.0463 | 0.2892 | 0.0081 | 0.2333 | 0.1414 | 0.3626 |
3.3072 | 6.0 | 612 | 1.6430 | 0.1371 | 0.301 | 0.1037 | 0.0347 | 0.1046 | 0.2527 | 0.1733 | 0.3374 | 0.3664 | 0.1605 | 0.3393 | 0.5973 | 0.4248 | 0.601 | 0.0216 | 0.2875 | 0.0572 | 0.3059 | 0.0243 | 0.2877 | 0.1577 | 0.35 |
3.3072 | 7.0 | 714 | 1.5879 | 0.1537 | 0.345 | 0.1218 | 0.0534 | 0.1358 | 0.2768 | 0.1933 | 0.3528 | 0.3807 | 0.1858 | 0.3554 | 0.6497 | 0.4326 | 0.6124 | 0.034 | 0.3486 | 0.0719 | 0.2917 | 0.0246 | 0.2877 | 0.2056 | 0.3631 |
3.3072 | 8.0 | 816 | 1.5310 | 0.1649 | 0.3587 | 0.1398 | 0.0579 | 0.1333 | 0.3036 | 0.1925 | 0.3729 | 0.3946 | 0.1854 | 0.3742 | 0.6476 | 0.4674 | 0.6353 | 0.0415 | 0.3722 | 0.0933 | 0.301 | 0.0387 | 0.3333 | 0.1836 | 0.3311 |
3.3072 | 9.0 | 918 | 1.4758 | 0.1789 | 0.3922 | 0.1478 | 0.0668 | 0.1372 | 0.3167 | 0.2241 | 0.3745 | 0.405 | 0.2063 | 0.3752 | 0.6691 | 0.4539 | 0.6199 | 0.063 | 0.4139 | 0.1041 | 0.3108 | 0.0375 | 0.3053 | 0.236 | 0.3752 |
1.4864 | 10.0 | 1020 | 1.4622 | 0.1735 | 0.3827 | 0.1333 | 0.05 | 0.1411 | 0.354 | 0.2103 | 0.371 | 0.3951 | 0.183 | 0.3664 | 0.6752 | 0.4784 | 0.6313 | 0.053 | 0.3903 | 0.1182 | 0.3186 | 0.0195 | 0.3053 | 0.1985 | 0.3301 |
1.4864 | 11.0 | 1122 | 1.4252 | 0.1858 | 0.4134 | 0.1496 | 0.0591 | 0.1632 | 0.3561 | 0.2227 | 0.3873 | 0.4144 | 0.1911 | 0.4137 | 0.6645 | 0.4794 | 0.6488 | 0.0752 | 0.4153 | 0.1131 | 0.3034 | 0.0292 | 0.3439 | 0.2319 | 0.3607 |
1.4864 | 12.0 | 1224 | 1.3893 | 0.1973 | 0.4218 | 0.1643 | 0.0749 | 0.169 | 0.4139 | 0.242 | 0.4054 | 0.4302 | 0.2226 | 0.4175 | 0.6991 | 0.4854 | 0.6413 | 0.0662 | 0.4292 | 0.1319 | 0.3397 | 0.0503 | 0.3561 | 0.2529 | 0.3845 |
1.4864 | 13.0 | 1326 | 1.3891 | 0.1998 | 0.431 | 0.1596 | 0.0675 | 0.1829 | 0.3762 | 0.2277 | 0.3962 | 0.4222 | 0.1979 | 0.4311 | 0.7011 | 0.504 | 0.6428 | 0.0911 | 0.4333 | 0.1384 | 0.3353 | 0.0552 | 0.3702 | 0.2101 | 0.3296 |
1.4864 | 14.0 | 1428 | 1.3981 | 0.193 | 0.42 | 0.1614 | 0.0698 | 0.1693 | 0.3523 | 0.235 | 0.3978 | 0.4271 | 0.2379 | 0.42 | 0.6729 | 0.4962 | 0.6557 | 0.0681 | 0.4278 | 0.136 | 0.3451 | 0.0493 | 0.3298 | 0.2155 | 0.3772 |
1.2306 | 15.0 | 1530 | 1.3472 | 0.217 | 0.4617 | 0.1785 | 0.0857 | 0.1817 | 0.4264 | 0.2416 | 0.4046 | 0.4329 | 0.2377 | 0.4143 | 0.7007 | 0.5137 | 0.6363 | 0.0968 | 0.4611 | 0.1571 | 0.3475 | 0.0484 | 0.3509 | 0.2689 | 0.3684 |
1.2306 | 16.0 | 1632 | 1.3450 | 0.227 | 0.4747 | 0.1915 | 0.0861 | 0.1891 | 0.4373 | 0.2521 | 0.4104 | 0.439 | 0.2503 | 0.4112 | 0.7344 | 0.5183 | 0.6428 | 0.1179 | 0.4514 | 0.1589 | 0.3289 | 0.0684 | 0.3912 | 0.2717 | 0.3806 |
1.2306 | 17.0 | 1734 | 1.2998 | 0.2359 | 0.4833 | 0.202 | 0.1089 | 0.1972 | 0.4426 | 0.2661 | 0.4303 | 0.4475 | 0.2792 | 0.4221 | 0.6999 | 0.5251 | 0.6463 | 0.12 | 0.4556 | 0.1646 | 0.3466 | 0.0857 | 0.393 | 0.284 | 0.3961 |
1.2306 | 18.0 | 1836 | 1.2995 | 0.2376 | 0.4866 | 0.1989 | 0.0926 | 0.2056 | 0.4487 | 0.2711 | 0.4325 | 0.4575 | 0.2798 | 0.4319 | 0.7195 | 0.522 | 0.6542 | 0.1299 | 0.475 | 0.1636 | 0.3544 | 0.0838 | 0.4018 | 0.2884 | 0.4019 |
1.2306 | 19.0 | 1938 | 1.2998 | 0.2362 | 0.4948 | 0.1954 | 0.1036 | 0.1905 | 0.4647 | 0.2563 | 0.4277 | 0.4446 | 0.249 | 0.4216 | 0.7165 | 0.5308 | 0.6672 | 0.1334 | 0.4722 | 0.1829 | 0.3407 | 0.0721 | 0.3772 | 0.2617 | 0.3655 |
1.0733 | 20.0 | 2040 | 1.2773 | 0.2513 | 0.5082 | 0.2298 | 0.1057 | 0.2148 | 0.4873 | 0.2723 | 0.4393 | 0.4678 | 0.2749 | 0.4514 | 0.7437 | 0.5342 | 0.6652 | 0.1499 | 0.4556 | 0.1754 | 0.3534 | 0.1101 | 0.4561 | 0.287 | 0.4087 |
1.0733 | 21.0 | 2142 | 1.2668 | 0.2516 | 0.5077 | 0.2323 | 0.1048 | 0.2104 | 0.4929 | 0.2758 | 0.4353 | 0.4592 | 0.2787 | 0.4287 | 0.7393 | 0.541 | 0.6692 | 0.1386 | 0.4653 | 0.1778 | 0.3525 | 0.1074 | 0.4018 | 0.2933 | 0.4073 |
1.0733 | 22.0 | 2244 | 1.2665 | 0.2496 | 0.5166 | 0.2143 | 0.114 | 0.2045 | 0.4759 | 0.2609 | 0.4314 | 0.454 | 0.2708 | 0.4246 | 0.7292 | 0.5355 | 0.6577 | 0.1393 | 0.4556 | 0.182 | 0.3657 | 0.1069 | 0.4 | 0.2842 | 0.3913 |
1.0733 | 23.0 | 2346 | 1.2512 | 0.2585 | 0.5258 | 0.2298 | 0.1196 | 0.2121 | 0.4884 | 0.2789 | 0.4465 | 0.4695 | 0.2991 | 0.4453 | 0.7262 | 0.5455 | 0.6672 | 0.1491 | 0.4833 | 0.1899 | 0.373 | 0.1149 | 0.4211 | 0.2931 | 0.4029 |
1.0733 | 24.0 | 2448 | 1.2511 | 0.2639 | 0.5275 | 0.2388 | 0.1198 | 0.2218 | 0.511 | 0.2845 | 0.4464 | 0.47 | 0.2911 | 0.4511 | 0.7377 | 0.5482 | 0.6657 | 0.1549 | 0.4694 | 0.192 | 0.3725 | 0.125 | 0.4386 | 0.2994 | 0.4039 |
0.9823 | 25.0 | 2550 | 1.2495 | 0.2629 | 0.5392 | 0.2309 | 0.1173 | 0.2213 | 0.4926 | 0.2828 | 0.4429 | 0.467 | 0.2888 | 0.4478 | 0.7363 | 0.549 | 0.6672 | 0.1633 | 0.4792 | 0.1931 | 0.3652 | 0.1181 | 0.4263 | 0.2908 | 0.3971 |
0.9823 | 26.0 | 2652 | 1.2470 | 0.2653 | 0.5276 | 0.2364 | 0.1136 | 0.2258 | 0.5082 | 0.2884 | 0.4486 | 0.4715 | 0.3017 | 0.4567 | 0.7313 | 0.5535 | 0.6701 | 0.1641 | 0.475 | 0.192 | 0.3672 | 0.1162 | 0.4368 | 0.3007 | 0.4083 |
0.9823 | 27.0 | 2754 | 1.2471 | 0.2661 | 0.5287 | 0.2366 | 0.1138 | 0.227 | 0.5013 | 0.2809 | 0.4483 | 0.4736 | 0.2986 | 0.4636 | 0.7286 | 0.5519 | 0.6711 | 0.1687 | 0.4806 | 0.1934 | 0.3676 | 0.1135 | 0.4404 | 0.3031 | 0.4083 |
0.9823 | 28.0 | 2856 | 1.2434 | 0.2673 | 0.5291 | 0.242 | 0.1156 | 0.229 | 0.5028 | 0.2866 | 0.4462 | 0.4745 | 0.3008 | 0.461 | 0.7367 | 0.5555 | 0.6736 | 0.1651 | 0.4806 | 0.1951 | 0.3662 | 0.1179 | 0.4421 | 0.3028 | 0.4102 |
0.9823 | 29.0 | 2958 | 1.2427 | 0.2676 | 0.5272 | 0.2425 | 0.116 | 0.2286 | 0.5 | 0.2863 | 0.4472 | 0.4745 | 0.299 | 0.4623 | 0.7343 | 0.554 | 0.6721 | 0.1675 | 0.4833 | 0.1942 | 0.3667 | 0.1195 | 0.4404 | 0.3027 | 0.4102 |
0.9316 | 30.0 | 3060 | 1.2426 | 0.268 | 0.5294 | 0.2419 | 0.1163 | 0.2288 | 0.5006 | 0.2865 | 0.4475 | 0.4749 | 0.3002 | 0.4623 | 0.7345 | 0.5546 | 0.6736 | 0.1674 | 0.4833 | 0.1944 | 0.3662 | 0.1199 | 0.4421 | 0.3036 | 0.4092 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.2.2+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 29
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for NotSarahConnor1984/detr_finetuned_cppe5
Base model
microsoft/conditional-detr-resnet-50