File size: 16,125 Bytes
ed6b6d1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 |
---
library_name: transformers
base_model: vladislavbro/dfine_l_coco
tags:
- generated_from_trainer
model-index:
- name: dfine_l_coco-cppe5-finetune
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# dfine_l_coco-cppe5-finetune
This model is a fine-tuned version of [vladislavbro/dfine_l_coco](https://huggingface.co/vladislavbro/dfine_l_coco) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.7496
- Map: 0.4281
- Map 50: 0.6128
- Map 75: 0.4751
- Map Small: 0.5217
- Map Medium: 0.3947
- Map Large: 0.5981
- Mar 1: 0.3348
- Mar 10: 0.6726
- Mar 100: 0.7561
- Mar Small: 0.6875
- Mar Medium: 0.6638
- Mar Large: 0.8701
- Map Coverall: 0.2303
- Mar 100 Coverall: 0.7923
- Map Face Shield: 0.559
- Mar 100 Face Shield: 0.7588
- Map Gloves: 0.5018
- Mar 100 Gloves: 0.7492
- Map Goggles: 0.3248
- Mar 100 Goggles: 0.7448
- Map Mask: 0.5244
- Mar 100 Mask: 0.7353
## 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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 300
- num_epochs: 40
### 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 | 12.2583 | 0.0076 | 0.017 | 0.0066 | 0.0004 | 0.002 | 0.0171 | 0.0283 | 0.0909 | 0.1254 | 0.0308 | 0.0555 | 0.2559 | 0.0322 | 0.295 | 0.0001 | 0.0506 | 0.0012 | 0.1147 | 0.0006 | 0.0846 | 0.004 | 0.0822 |
| No log | 2.0 | 214 | 6.2711 | 0.0583 | 0.1115 | 0.0522 | 0.0223 | 0.0361 | 0.09 | 0.092 | 0.2671 | 0.3486 | 0.1993 | 0.2602 | 0.568 | 0.1623 | 0.4869 | 0.0048 | 0.3139 | 0.0281 | 0.2795 | 0.0119 | 0.28 | 0.0844 | 0.3827 |
| No log | 3.0 | 321 | 4.7100 | 0.087 | 0.1571 | 0.083 | 0.041 | 0.0909 | 0.1375 | 0.1294 | 0.3482 | 0.4407 | 0.29 | 0.351 | 0.6387 | 0.1513 | 0.6532 | 0.0134 | 0.4076 | 0.0659 | 0.333 | 0.0127 | 0.3908 | 0.1916 | 0.4191 |
| No log | 4.0 | 428 | 4.2788 | 0.1351 | 0.2495 | 0.1295 | 0.0714 | 0.122 | 0.2464 | 0.1838 | 0.3833 | 0.4827 | 0.3506 | 0.3924 | 0.6756 | 0.2092 | 0.6293 | 0.0668 | 0.4696 | 0.1164 | 0.4433 | 0.0349 | 0.4046 | 0.2481 | 0.4667 |
| 11.3943 | 5.0 | 535 | 4.0960 | 0.1555 | 0.2785 | 0.1545 | 0.128 | 0.1607 | 0.252 | 0.1898 | 0.416 | 0.5219 | 0.355 | 0.4384 | 0.7168 | 0.1857 | 0.6329 | 0.1135 | 0.5392 | 0.1512 | 0.475 | 0.0836 | 0.4738 | 0.2435 | 0.4884 |
| 11.3943 | 6.0 | 642 | 4.0103 | 0.1497 | 0.2754 | 0.1512 | 0.0884 | 0.1426 | 0.2863 | 0.2146 | 0.4285 | 0.5291 | 0.3729 | 0.4504 | 0.6933 | 0.2234 | 0.6653 | 0.0682 | 0.557 | 0.1849 | 0.4509 | 0.065 | 0.5046 | 0.2071 | 0.4676 |
| 11.3943 | 7.0 | 749 | 3.7960 | 0.1528 | 0.2793 | 0.1543 | 0.0742 | 0.1214 | 0.3125 | 0.2204 | 0.4128 | 0.5311 | 0.3611 | 0.4468 | 0.7385 | 0.2321 | 0.686 | 0.078 | 0.5139 | 0.1677 | 0.4862 | 0.059 | 0.4662 | 0.227 | 0.5031 |
| 11.3943 | 8.0 | 856 | 3.7461 | 0.1474 | 0.2686 | 0.1466 | 0.0955 | 0.1477 | 0.2527 | 0.2114 | 0.4098 | 0.5052 | 0.322 | 0.4119 | 0.6795 | 0.2048 | 0.6473 | 0.0694 | 0.4418 | 0.1723 | 0.4705 | 0.0286 | 0.4785 | 0.2622 | 0.488 |
| 11.3943 | 9.0 | 963 | 3.6575 | 0.1386 | 0.2621 | 0.1295 | 0.1228 | 0.1357 | 0.23 | 0.1948 | 0.4493 | 0.5514 | 0.3762 | 0.5058 | 0.7323 | 0.1968 | 0.7077 | 0.0786 | 0.5468 | 0.1729 | 0.5022 | 0.0351 | 0.4923 | 0.2099 | 0.508 |
| 4.7509 | 10.0 | 1070 | 3.6422 | 0.1395 | 0.27 | 0.1287 | 0.1266 | 0.1418 | 0.249 | 0.201 | 0.4301 | 0.5342 | 0.3716 | 0.4432 | 0.7413 | 0.1311 | 0.6797 | 0.1029 | 0.5089 | 0.1867 | 0.4884 | 0.0799 | 0.4892 | 0.1971 | 0.5049 |
| 4.7509 | 11.0 | 1177 | 3.5627 | 0.1822 | 0.3267 | 0.1794 | 0.1295 | 0.1738 | 0.309 | 0.2261 | 0.4396 | 0.5464 | 0.3643 | 0.4763 | 0.7369 | 0.2486 | 0.7045 | 0.1213 | 0.5253 | 0.2195 | 0.5085 | 0.0906 | 0.5 | 0.231 | 0.4938 |
| 4.7509 | 12.0 | 1284 | 3.5589 | 0.1776 | 0.322 | 0.1704 | 0.1469 | 0.1596 | 0.2852 | 0.2269 | 0.4563 | 0.5585 | 0.4086 | 0.4839 | 0.7538 | 0.2413 | 0.6946 | 0.1428 | 0.5405 | 0.2158 | 0.5254 | 0.0761 | 0.5308 | 0.212 | 0.5013 |
| 4.7509 | 13.0 | 1391 | 3.5184 | 0.1667 | 0.3119 | 0.1545 | 0.1451 | 0.1796 | 0.2856 | 0.2249 | 0.4335 | 0.5406 | 0.3279 | 0.4512 | 0.7274 | 0.1494 | 0.6608 | 0.1201 | 0.5038 | 0.226 | 0.5129 | 0.0864 | 0.5231 | 0.2516 | 0.5022 |
| 4.7509 | 14.0 | 1498 | 3.4570 | 0.2238 | 0.3985 | 0.2242 | 0.1642 | 0.1924 | 0.3459 | 0.2632 | 0.4688 | 0.5615 | 0.4281 | 0.5003 | 0.7426 | 0.2959 | 0.6986 | 0.2161 | 0.5835 | 0.2167 | 0.5 | 0.0976 | 0.5231 | 0.2925 | 0.5022 |
| 4.2981 | 15.0 | 1605 | 3.4975 | 0.1898 | 0.3529 | 0.1835 | 0.1699 | 0.1706 | 0.3331 | 0.2261 | 0.4625 | 0.5597 | 0.4369 | 0.477 | 0.7396 | 0.2577 | 0.7027 | 0.1725 | 0.5456 | 0.2175 | 0.5317 | 0.059 | 0.5185 | 0.2421 | 0.5 |
| 4.2981 | 16.0 | 1712 | 3.4823 | 0.2029 | 0.3588 | 0.1989 | 0.185 | 0.1916 | 0.3601 | 0.237 | 0.4557 | 0.5619 | 0.4026 | 0.4934 | 0.7379 | 0.1876 | 0.6865 | 0.1993 | 0.5506 | 0.2333 | 0.5344 | 0.1213 | 0.5446 | 0.2729 | 0.4933 |
| 4.2981 | 17.0 | 1819 | 3.4680 | 0.1966 | 0.3431 | 0.1914 | 0.1928 | 0.2102 | 0.3505 | 0.2367 | 0.4482 | 0.5614 | 0.3739 | 0.4811 | 0.7557 | 0.1899 | 0.6878 | 0.1736 | 0.5253 | 0.229 | 0.5348 | 0.1088 | 0.5569 | 0.2818 | 0.5022 |
| 4.2981 | 18.0 | 1926 | 3.5169 | 0.1921 | 0.3388 | 0.1911 | 0.1347 | 0.1813 | 0.3905 | 0.2223 | 0.4443 | 0.5567 | 0.3799 | 0.4839 | 0.7512 | 0.1424 | 0.6617 | 0.168 | 0.5253 | 0.2432 | 0.5571 | 0.1306 | 0.52 | 0.2761 | 0.5196 |
| 4.0614 | 19.0 | 2033 | 3.4963 | 0.2067 | 0.365 | 0.1956 | 0.139 | 0.1892 | 0.3829 | 0.2504 | 0.4607 | 0.5676 | 0.4202 | 0.4924 | 0.7511 | 0.1461 | 0.682 | 0.2216 | 0.5722 | 0.2319 | 0.5482 | 0.1713 | 0.5323 | 0.2628 | 0.5031 |
| 4.0614 | 20.0 | 2140 | 3.5090 | 0.1924 | 0.3403 | 0.1858 | 0.1192 | 0.1871 | 0.3734 | 0.2191 | 0.4486 | 0.5493 | 0.3453 | 0.4729 | 0.748 | 0.0925 | 0.6815 | 0.1988 | 0.5013 | 0.2504 | 0.5585 | 0.1529 | 0.5108 | 0.2672 | 0.4942 |
| 4.0614 | 21.0 | 2247 | 3.4683 | 0.1993 | 0.3635 | 0.1858 | 0.145 | 0.1918 | 0.371 | 0.2372 | 0.4411 | 0.5551 | 0.3925 | 0.484 | 0.7343 | 0.13 | 0.6883 | 0.1896 | 0.5304 | 0.2478 | 0.542 | 0.1657 | 0.5031 | 0.2633 | 0.5116 |
| 4.0614 | 22.0 | 2354 | 3.4556 | 0.2066 | 0.3627 | 0.2075 | 0.1562 | 0.1901 | 0.3959 | 0.2452 | 0.4486 | 0.553 | 0.3986 | 0.477 | 0.7445 | 0.1576 | 0.6874 | 0.2169 | 0.5595 | 0.252 | 0.5263 | 0.1653 | 0.4846 | 0.2413 | 0.5071 |
| 4.0614 | 23.0 | 2461 | 3.4420 | 0.2009 | 0.3516 | 0.1954 | 0.18 | 0.1826 | 0.4079 | 0.2378 | 0.4519 | 0.5514 | 0.3677 | 0.4749 | 0.7416 | 0.1585 | 0.6919 | 0.1463 | 0.5342 | 0.2631 | 0.5263 | 0.1687 | 0.4969 | 0.2678 | 0.5076 |
| 3.8839 | 24.0 | 2568 | 3.3910 | 0.2333 | 0.4059 | 0.2262 | 0.1423 | 0.1911 | 0.4386 | 0.2506 | 0.4617 | 0.5561 | 0.3432 | 0.486 | 0.7486 | 0.2541 | 0.7036 | 0.1829 | 0.5304 | 0.2639 | 0.542 | 0.1909 | 0.4954 | 0.2748 | 0.5093 |
| 3.8839 | 25.0 | 2675 | 3.4301 | 0.222 | 0.3794 | 0.2246 | 0.152 | 0.1998 | 0.3831 | 0.245 | 0.4505 | 0.5346 | 0.3242 | 0.4696 | 0.722 | 0.2289 | 0.6932 | 0.1822 | 0.4873 | 0.2853 | 0.5554 | 0.1424 | 0.4431 | 0.2713 | 0.4938 |
| 3.8839 | 26.0 | 2782 | 3.4264 | 0.203 | 0.3622 | 0.1955 | 0.1845 | 0.1811 | 0.388 | 0.2422 | 0.4466 | 0.5467 | 0.3682 | 0.4734 | 0.7354 | 0.1886 | 0.6874 | 0.1586 | 0.4886 | 0.2498 | 0.5348 | 0.1347 | 0.52 | 0.2834 | 0.5027 |
| 3.8839 | 27.0 | 2889 | 3.3817 | 0.2254 | 0.3928 | 0.2197 | 0.203 | 0.1945 | 0.4092 | 0.2524 | 0.4532 | 0.5564 | 0.3589 | 0.4826 | 0.74 | 0.1935 | 0.6806 | 0.1759 | 0.5215 | 0.2837 | 0.5513 | 0.182 | 0.5185 | 0.2919 | 0.5102 |
| 3.8839 | 28.0 | 2996 | 3.4003 | 0.2256 | 0.3852 | 0.2182 | 0.1903 | 0.2088 | 0.4035 | 0.2552 | 0.4636 | 0.5604 | 0.3932 | 0.4856 | 0.7376 | 0.146 | 0.6932 | 0.2443 | 0.5468 | 0.2793 | 0.5509 | 0.1566 | 0.5046 | 0.302 | 0.5062 |
| 3.7638 | 29.0 | 3103 | 3.3650 | 0.2342 | 0.4079 | 0.2297 | 0.1691 | 0.2007 | 0.4231 | 0.2638 | 0.4664 | 0.5654 | 0.3828 | 0.4885 | 0.7596 | 0.2185 | 0.7063 | 0.2383 | 0.5608 | 0.2726 | 0.5558 | 0.1471 | 0.5046 | 0.2948 | 0.4996 |
| 3.7638 | 30.0 | 3210 | 3.3953 | 0.2297 | 0.4083 | 0.2263 | 0.1908 | 0.2014 | 0.4143 | 0.2484 | 0.4648 | 0.5632 | 0.3895 | 0.4812 | 0.7554 | 0.1799 | 0.705 | 0.2465 | 0.5481 | 0.2694 | 0.5594 | 0.1581 | 0.4892 | 0.2944 | 0.5142 |
| 3.7638 | 31.0 | 3317 | 3.3695 | 0.2321 | 0.4052 | 0.2357 | 0.1867 | 0.2086 | 0.4357 | 0.2659 | 0.4676 | 0.5651 | 0.39 | 0.491 | 0.7463 | 0.1686 | 0.6937 | 0.2441 | 0.5519 | 0.2729 | 0.5384 | 0.1775 | 0.5354 | 0.2975 | 0.5062 |
| 3.7638 | 32.0 | 3424 | 3.3628 | 0.2376 | 0.4163 | 0.2363 | 0.1931 | 0.2083 | 0.4274 | 0.2681 | 0.4661 | 0.5522 | 0.354 | 0.4712 | 0.7547 | 0.1974 | 0.6995 | 0.2428 | 0.5127 | 0.2792 | 0.5397 | 0.1609 | 0.4954 | 0.3077 | 0.5138 |
| 3.6477 | 33.0 | 3531 | 3.3567 | 0.2392 | 0.4142 | 0.2331 | 0.2006 | 0.2027 | 0.4335 | 0.2702 | 0.4717 | 0.5703 | 0.3965 | 0.4984 | 0.7471 | 0.1983 | 0.695 | 0.2472 | 0.5633 | 0.275 | 0.5629 | 0.1709 | 0.5169 | 0.3045 | 0.5133 |
| 3.6477 | 34.0 | 3638 | 3.3581 | 0.2357 | 0.4126 | 0.2298 | 0.1897 | 0.2126 | 0.4235 | 0.2594 | 0.4593 | 0.5509 | 0.3477 | 0.473 | 0.7392 | 0.178 | 0.6932 | 0.2369 | 0.5076 | 0.2834 | 0.5571 | 0.1692 | 0.48 | 0.311 | 0.5164 |
| 3.6477 | 35.0 | 3745 | 3.3458 | 0.2396 | 0.4177 | 0.2337 | 0.1989 | 0.2046 | 0.4346 | 0.254 | 0.4651 | 0.5626 | 0.3517 | 0.493 | 0.7448 | 0.2062 | 0.709 | 0.2244 | 0.5228 | 0.2838 | 0.5625 | 0.1758 | 0.5092 | 0.308 | 0.5093 |
| 3.6477 | 36.0 | 3852 | 3.3376 | 0.2385 | 0.419 | 0.2413 | 0.192 | 0.2111 | 0.4405 | 0.25 | 0.4648 | 0.5564 | 0.3456 | 0.4818 | 0.7521 | 0.1978 | 0.7009 | 0.2216 | 0.5063 | 0.2831 | 0.5491 | 0.1734 | 0.5154 | 0.3167 | 0.5102 |
| 3.6477 | 37.0 | 3959 | 3.3495 | 0.2346 | 0.4103 | 0.2254 | 0.2041 | 0.211 | 0.4315 | 0.2689 | 0.4662 | 0.5579 | 0.3562 | 0.4948 | 0.7437 | 0.1723 | 0.6833 | 0.2407 | 0.5278 | 0.2739 | 0.5487 | 0.1757 | 0.5169 | 0.3106 | 0.5129 |
| 3.5391 | 38.0 | 4066 | 3.3427 | 0.2369 | 0.4094 | 0.2328 | 0.1985 | 0.2118 | 0.4304 | 0.2582 | 0.4669 | 0.5656 | 0.3609 | 0.4957 | 0.7469 | 0.174 | 0.7041 | 0.24 | 0.538 | 0.2738 | 0.5522 | 0.1868 | 0.5215 | 0.3097 | 0.5124 |
| 3.5391 | 39.0 | 4173 | 3.3297 | 0.2371 | 0.4149 | 0.2301 | 0.1992 | 0.2096 | 0.4322 | 0.2619 | 0.4668 | 0.5626 | 0.3571 | 0.493 | 0.7464 | 0.1956 | 0.7005 | 0.2222 | 0.5253 | 0.2832 | 0.5661 | 0.1759 | 0.5077 | 0.3089 | 0.5133 |
| 3.5391 | 40.0 | 4280 | 3.3385 | 0.2371 | 0.4143 | 0.2309 | 0.1937 | 0.211 | 0.4367 | 0.2604 | 0.4634 | 0.5594 | 0.3529 | 0.4907 | 0.7388 | 0.1874 | 0.6986 | 0.2384 | 0.5241 | 0.2805 | 0.5652 | 0.1742 | 0.5 | 0.3048 | 0.5093 |
### Framework versions
- Transformers 4.50.0.dev0
- Pytorch 2.6.0+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0
|