--- license: apache-2.0 base_model: microsoft/conditional-detr-resnet-50 tags: - generated_from_trainer model-index: - name: detr_finetuned_cppe5 results: [] --- # detr_finetuned_cppe5 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. It achieves the following results on the evaluation set: - Loss: 1.3407 - Map: 0.2599 - Map 50: 0.5107 - Map 75: 0.2411 - Map Small: 0.1265 - Map Medium: 0.2152 - Map Large: 0.4809 - Mar 1: 0.2669 - Mar 10: 0.4141 - Mar 100: 0.4315 - Mar Small: 0.2471 - Mar Medium: 0.4009 - Mar Large: 0.7004 - Map Coverall: 0.5407 - Mar 100 Coverall: 0.6477 - Map Face Shield: 0.1688 - Mar 100 Face Shield: 0.4532 - Map Gloves: 0.1974 - Mar 100 Gloves: 0.3344 - Map Goggles: 0.1266 - Mar 100 Goggles: 0.3415 - Map Mask: 0.266 - Mar 100 Mask: 0.3804 ## 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 | 107 | 2.4182 | 0.0507 | 0.104 | 0.0439 | 0.0022 | 0.0243 | 0.0555 | 0.0607 | 0.1378 | 0.1804 | 0.028 | 0.1513 | 0.2649 | 0.2397 | 0.5194 | 0.0003 | 0.0494 | 0.0029 | 0.0969 | 0.0 | 0.0 | 0.0105 | 0.2364 | | No log | 2.0 | 214 | 2.1888 | 0.0484 | 0.0991 | 0.0426 | 0.0128 | 0.0264 | 0.0477 | 0.0773 | 0.1617 | 0.2023 | 0.0433 | 0.1502 | 0.2651 | 0.1982 | 0.5892 | 0.0001 | 0.0101 | 0.0168 | 0.1625 | 0.0015 | 0.0108 | 0.0255 | 0.2387 | | No log | 3.0 | 321 | 2.0106 | 0.0827 | 0.1666 | 0.0735 | 0.0148 | 0.0543 | 0.1059 | 0.109 | 0.2402 | 0.2787 | 0.0696 | 0.2816 | 0.3968 | 0.304 | 0.6144 | 0.0053 | 0.1671 | 0.0206 | 0.2455 | 0.0118 | 0.0385 | 0.072 | 0.328 | | No log | 4.0 | 428 | 1.9302 | 0.107 | 0.2298 | 0.0892 | 0.0258 | 0.0669 | 0.1511 | 0.1338 | 0.2939 | 0.3207 | 0.1213 | 0.2985 | 0.4868 | 0.3695 | 0.5797 | 0.016 | 0.2911 | 0.0302 | 0.2464 | 0.0128 | 0.16 | 0.1066 | 0.3262 | | 3.6586 | 5.0 | 535 | 1.8116 | 0.1183 | 0.2773 | 0.0879 | 0.0292 | 0.0782 | 0.1819 | 0.1467 | 0.3143 | 0.3378 | 0.142 | 0.3138 | 0.5634 | 0.3744 | 0.5658 | 0.052 | 0.3494 | 0.0532 | 0.2652 | 0.007 | 0.1769 | 0.1049 | 0.332 | | 3.6586 | 6.0 | 642 | 1.7759 | 0.1213 | 0.2878 | 0.0867 | 0.019 | 0.0851 | 0.2366 | 0.1369 | 0.3103 | 0.3372 | 0.1215 | 0.3085 | 0.6211 | 0.4062 | 0.5694 | 0.0278 | 0.3278 | 0.0582 | 0.2594 | 0.0128 | 0.2338 | 0.1017 | 0.2956 | | 3.6586 | 7.0 | 749 | 1.6378 | 0.1555 | 0.3462 | 0.1182 | 0.0467 | 0.1064 | 0.2873 | 0.168 | 0.3436 | 0.3788 | 0.1635 | 0.3456 | 0.6683 | 0.4273 | 0.5658 | 0.0479 | 0.3873 | 0.0927 | 0.3022 | 0.0277 | 0.2908 | 0.1819 | 0.348 | | 3.6586 | 8.0 | 856 | 1.6132 | 0.1654 | 0.376 | 0.1226 | 0.0495 | 0.1358 | 0.366 | 0.1966 | 0.353 | 0.3824 | 0.1421 | 0.3667 | 0.6958 | 0.4324 | 0.5833 | 0.0803 | 0.4152 | 0.1005 | 0.2853 | 0.0373 | 0.2754 | 0.1766 | 0.3529 | | 3.6586 | 9.0 | 963 | 1.5567 | 0.1815 | 0.3979 | 0.1439 | 0.0529 | 0.1518 | 0.3407 | 0.2063 | 0.3721 | 0.396 | 0.1506 | 0.3879 | 0.6784 | 0.4654 | 0.6149 | 0.0841 | 0.4063 | 0.1158 | 0.3058 | 0.041 | 0.3015 | 0.201 | 0.3516 | | 1.5229 | 10.0 | 1070 | 1.5420 | 0.194 | 0.4056 | 0.1562 | 0.0523 | 0.1635 | 0.3805 | 0.2139 | 0.3706 | 0.3952 | 0.1409 | 0.3849 | 0.7155 | 0.4799 | 0.6338 | 0.1029 | 0.4114 | 0.127 | 0.3089 | 0.0401 | 0.2754 | 0.22 | 0.3467 | | 1.5229 | 11.0 | 1177 | 1.4853 | 0.2006 | 0.4273 | 0.1683 | 0.0753 | 0.1676 | 0.3949 | 0.2214 | 0.3753 | 0.4054 | 0.18 | 0.3976 | 0.6702 | 0.4916 | 0.6167 | 0.1162 | 0.4241 | 0.1199 | 0.3138 | 0.0464 | 0.3185 | 0.229 | 0.3542 | | 1.5229 | 12.0 | 1284 | 1.4646 | 0.2054 | 0.4336 | 0.1626 | 0.0809 | 0.1636 | 0.4116 | 0.2244 | 0.3933 | 0.4162 | 0.196 | 0.4011 | 0.688 | 0.4921 | 0.6333 | 0.098 | 0.4291 | 0.152 | 0.2991 | 0.0556 | 0.3692 | 0.2293 | 0.3502 | | 1.5229 | 13.0 | 1391 | 1.4438 | 0.2113 | 0.4421 | 0.176 | 0.0722 | 0.1721 | 0.4278 | 0.2333 | 0.3903 | 0.4108 | 0.1905 | 0.4006 | 0.6807 | 0.5002 | 0.6315 | 0.1082 | 0.4342 | 0.1602 | 0.3085 | 0.0488 | 0.3169 | 0.2391 | 0.3627 | | 1.5229 | 14.0 | 1498 | 1.4194 | 0.2241 | 0.4597 | 0.1846 | 0.0857 | 0.1878 | 0.4516 | 0.2418 | 0.3973 | 0.4209 | 0.1983 | 0.4126 | 0.7007 | 0.5049 | 0.6104 | 0.1265 | 0.4291 | 0.1644 | 0.3299 | 0.0686 | 0.3569 | 0.2564 | 0.3782 | | 1.2614 | 15.0 | 1605 | 1.4168 | 0.2194 | 0.4409 | 0.191 | 0.0921 | 0.172 | 0.443 | 0.2416 | 0.3979 | 0.4213 | 0.2283 | 0.39 | 0.6824 | 0.5237 | 0.6441 | 0.1208 | 0.4557 | 0.1581 | 0.3129 | 0.0595 | 0.3246 | 0.235 | 0.3689 | | 1.2614 | 16.0 | 1712 | 1.3935 | 0.226 | 0.4735 | 0.187 | 0.0995 | 0.1831 | 0.4229 | 0.237 | 0.4015 | 0.4238 | 0.2175 | 0.4082 | 0.6808 | 0.5125 | 0.6288 | 0.1292 | 0.4734 | 0.1735 | 0.3263 | 0.0566 | 0.32 | 0.2584 | 0.3702 | | 1.2614 | 17.0 | 1819 | 1.3928 | 0.2295 | 0.4823 | 0.1949 | 0.0841 | 0.1911 | 0.441 | 0.2507 | 0.3996 | 0.4201 | 0.2206 | 0.3903 | 0.7086 | 0.5135 | 0.632 | 0.1465 | 0.4557 | 0.1652 | 0.3246 | 0.0767 | 0.3169 | 0.2456 | 0.3716 | | 1.2614 | 18.0 | 1926 | 1.3886 | 0.2302 | 0.4745 | 0.1908 | 0.0836 | 0.1922 | 0.4742 | 0.2562 | 0.404 | 0.4203 | 0.199 | 0.3884 | 0.7143 | 0.5158 | 0.6347 | 0.1484 | 0.4582 | 0.1736 | 0.3192 | 0.064 | 0.3215 | 0.2491 | 0.368 | | 1.104 | 19.0 | 2033 | 1.3812 | 0.2343 | 0.4775 | 0.201 | 0.0954 | 0.1982 | 0.4586 | 0.248 | 0.3985 | 0.4221 | 0.2093 | 0.4013 | 0.7229 | 0.5257 | 0.641 | 0.1555 | 0.462 | 0.1778 | 0.3308 | 0.0791 | 0.32 | 0.2336 | 0.3569 | | 1.104 | 20.0 | 2140 | 1.3595 | 0.2488 | 0.4941 | 0.2209 | 0.0973 | 0.2065 | 0.4771 | 0.2677 | 0.4188 | 0.4369 | 0.2404 | 0.4026 | 0.7248 | 0.5337 | 0.6441 | 0.1672 | 0.4709 | 0.1832 | 0.3335 | 0.094 | 0.3523 | 0.2658 | 0.3836 | | 1.104 | 21.0 | 2247 | 1.3556 | 0.2397 | 0.4789 | 0.2046 | 0.0941 | 0.1986 | 0.4552 | 0.2683 | 0.4094 | 0.4298 | 0.2244 | 0.4045 | 0.7063 | 0.5311 | 0.6396 | 0.1483 | 0.4506 | 0.1868 | 0.3304 | 0.0785 | 0.3508 | 0.2537 | 0.3778 | | 1.104 | 22.0 | 2354 | 1.3572 | 0.2509 | 0.4949 | 0.2242 | 0.1067 | 0.2086 | 0.4512 | 0.2672 | 0.4119 | 0.4308 | 0.2403 | 0.397 | 0.7136 | 0.5405 | 0.6432 | 0.1641 | 0.4595 | 0.176 | 0.3237 | 0.1085 | 0.3431 | 0.2653 | 0.3844 | | 1.104 | 23.0 | 2461 | 1.3551 | 0.2503 | 0.4951 | 0.2266 | 0.1053 | 0.2057 | 0.476 | 0.2674 | 0.4117 | 0.4297 | 0.2319 | 0.4058 | 0.7042 | 0.5403 | 0.6464 | 0.1522 | 0.4367 | 0.1828 | 0.3299 | 0.1129 | 0.3508 | 0.2633 | 0.3849 | | 1.0066 | 24.0 | 2568 | 1.3404 | 0.2539 | 0.5049 | 0.2235 | 0.101 | 0.2081 | 0.4745 | 0.2674 | 0.412 | 0.4301 | 0.231 | 0.4038 | 0.692 | 0.5437 | 0.6559 | 0.1537 | 0.4367 | 0.1903 | 0.3348 | 0.1218 | 0.3415 | 0.2601 | 0.3813 | | 1.0066 | 25.0 | 2675 | 1.3436 | 0.2574 | 0.5062 | 0.2286 | 0.1124 | 0.2119 | 0.4848 | 0.2667 | 0.4101 | 0.4273 | 0.2264 | 0.4014 | 0.6942 | 0.5416 | 0.6477 | 0.1512 | 0.4329 | 0.193 | 0.3366 | 0.131 | 0.3369 | 0.2702 | 0.3822 | | 1.0066 | 26.0 | 2782 | 1.3377 | 0.258 | 0.5047 | 0.2211 | 0.1254 | 0.2126 | 0.4825 | 0.27 | 0.4168 | 0.4348 | 0.2491 | 0.4062 | 0.7013 | 0.5431 | 0.6518 | 0.1604 | 0.462 | 0.1935 | 0.3397 | 0.1259 | 0.34 | 0.2669 | 0.3804 | | 1.0066 | 27.0 | 2889 | 1.3393 | 0.2615 | 0.5108 | 0.2388 | 0.1277 | 0.2188 | 0.4796 | 0.2711 | 0.4167 | 0.4347 | 0.2509 | 0.408 | 0.6993 | 0.5427 | 0.6491 | 0.1685 | 0.4608 | 0.1949 | 0.3348 | 0.1315 | 0.3462 | 0.2699 | 0.3827 | | 1.0066 | 28.0 | 2996 | 1.3399 | 0.2599 | 0.5102 | 0.2352 | 0.1259 | 0.2166 | 0.4843 | 0.2674 | 0.415 | 0.4326 | 0.2482 | 0.4012 | 0.7042 | 0.5419 | 0.65 | 0.1678 | 0.4544 | 0.1945 | 0.3357 | 0.1253 | 0.3385 | 0.2698 | 0.3844 | | 0.95 | 29.0 | 3103 | 1.3412 | 0.2594 | 0.5122 | 0.2387 | 0.1259 | 0.2159 | 0.4808 | 0.2702 | 0.4143 | 0.4303 | 0.2452 | 0.3983 | 0.7016 | 0.5393 | 0.6468 | 0.1689 | 0.4532 | 0.197 | 0.3335 | 0.1252 | 0.3369 | 0.2667 | 0.3809 | | 0.95 | 30.0 | 3210 | 1.3407 | 0.2599 | 0.5107 | 0.2411 | 0.1265 | 0.2152 | 0.4809 | 0.2669 | 0.4141 | 0.4315 | 0.2471 | 0.4009 | 0.7004 | 0.5407 | 0.6477 | 0.1688 | 0.4532 | 0.1974 | 0.3344 | 0.1266 | 0.3415 | 0.266 | 0.3804 | ### Framework versions - Transformers 4.40.2 - Pytorch 2.2.1+cu121 - Datasets 2.19.1 - Tokenizers 0.19.1