metadata
base_model: PekingU/rtdetr_r50vd_coco_o365
tags:
- generated_from_trainer
model-index:
- name: rtdetr-r50-cppe5-finetune-v3
results: []
rtdetr-r50-cppe5-finetune-v3
This model is a fine-tuned version of PekingU/rtdetr_r50vd_coco_o365 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 10.9472
- Map: 0.3704
- Map 50: 0.5798
- Map 75: 0.3751
- Map Small: 0.1898
- Map Medium: 0.3794
- Map Large: 0.4798
- Mar 1: 0.3002
- Mar 10: 0.5262
- Mar 100: 0.6137
- Mar Small: 0.4445
- Mar Medium: 0.5564
- Mar Large: 0.8118
- Map Coverall: 0.4594
- Mar 100 Coverall: 0.6795
- Map Face Shield: 0.4864
- Mar 100 Face Shield: 0.6412
- Map Gloves: 0.3727
- Mar 100 Gloves: 0.6322
- Map Goggles: 0.1661
- Mar 100 Goggles: 0.4586
- Map Mask: 0.3674
- Mar 100 Mask: 0.6569
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: linear
- num_epochs: 30
- mixed_precision_training: Native AMP
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 | 113.2495 | 0.011 | 0.0228 | 0.0089 | 0.0002 | 0.0063 | 0.0248 | 0.0285 | 0.1032 | 0.1933 | 0.0311 | 0.1471 | 0.3682 | 0.053 | 0.5482 | 0.0002 | 0.1316 | 0.0001 | 0.0379 | 0.0 | 0.0169 | 0.0016 | 0.232 |
No log | 2.0 | 214 | 18.6938 | 0.1584 | 0.2939 | 0.1442 | 0.0773 | 0.1147 | 0.2221 | 0.1645 | 0.3785 | 0.4702 | 0.303 | 0.4358 | 0.7243 | 0.4389 | 0.6694 | 0.0065 | 0.4266 | 0.0819 | 0.4272 | 0.0187 | 0.3585 | 0.2462 | 0.4693 |
No log | 3.0 | 321 | 12.9839 | 0.226 | 0.3923 | 0.2176 | 0.1203 | 0.1804 | 0.4483 | 0.2371 | 0.427 | 0.5082 | 0.3272 | 0.4915 | 0.7484 | 0.4397 | 0.668 | 0.0681 | 0.4608 | 0.1557 | 0.4585 | 0.1696 | 0.4662 | 0.2969 | 0.4876 |
No log | 4.0 | 428 | 12.4463 | 0.215 | 0.4188 | 0.1966 | 0.1012 | 0.1805 | 0.4172 | 0.2229 | 0.4095 | 0.4812 | 0.2833 | 0.4737 | 0.6995 | 0.3622 | 0.6716 | 0.1302 | 0.4975 | 0.1221 | 0.4205 | 0.1687 | 0.3754 | 0.2916 | 0.4409 |
83.7659 | 5.0 | 535 | 12.3391 | 0.2336 | 0.465 | 0.2029 | 0.1382 | 0.2226 | 0.5186 | 0.2355 | 0.4423 | 0.5212 | 0.3433 | 0.5183 | 0.7393 | 0.2065 | 0.6703 | 0.2304 | 0.5051 | 0.1915 | 0.45 | 0.2376 | 0.4785 | 0.3021 | 0.5022 |
83.7659 | 6.0 | 642 | 12.3727 | 0.2059 | 0.3923 | 0.1857 | 0.1211 | 0.2121 | 0.4664 | 0.2274 | 0.4292 | 0.5156 | 0.3612 | 0.504 | 0.7148 | 0.1794 | 0.6694 | 0.2126 | 0.5329 | 0.1643 | 0.4455 | 0.1688 | 0.42 | 0.3042 | 0.5102 |
83.7659 | 7.0 | 749 | 12.7240 | 0.2058 | 0.41 | 0.1759 | 0.1217 | 0.2158 | 0.4525 | 0.2295 | 0.4086 | 0.4943 | 0.3464 | 0.4782 | 0.7071 | 0.2036 | 0.6077 | 0.2208 | 0.5013 | 0.1332 | 0.4201 | 0.2176 | 0.4323 | 0.2539 | 0.5102 |
83.7659 | 8.0 | 856 | 13.0090 | 0.2015 | 0.3831 | 0.1849 | 0.1056 | 0.2061 | 0.4551 | 0.2227 | 0.4194 | 0.5332 | 0.3719 | 0.5245 | 0.7451 | 0.2041 | 0.6482 | 0.1889 | 0.5747 | 0.1627 | 0.4692 | 0.1792 | 0.4462 | 0.2727 | 0.5276 |
83.7659 | 9.0 | 963 | 12.9929 | 0.2136 | 0.4059 | 0.1897 | 0.1307 | 0.1946 | 0.4823 | 0.2372 | 0.4132 | 0.5178 | 0.3559 | 0.5187 | 0.7195 | 0.2503 | 0.6248 | 0.22 | 0.5177 | 0.2002 | 0.4799 | 0.1888 | 0.46 | 0.2088 | 0.5067 |
12.788 | 10.0 | 1070 | 13.2438 | 0.1824 | 0.3375 | 0.1626 | 0.1 | 0.1911 | 0.4538 | 0.2061 | 0.3935 | 0.5026 | 0.3561 | 0.5017 | 0.6963 | 0.176 | 0.5644 | 0.1629 | 0.5228 | 0.1649 | 0.4741 | 0.1767 | 0.4769 | 0.2312 | 0.4747 |
12.788 | 11.0 | 1177 | 13.0890 | 0.2074 | 0.3885 | 0.1917 | 0.0937 | 0.1761 | 0.4555 | 0.2372 | 0.4106 | 0.508 | 0.3414 | 0.4956 | 0.7394 | 0.353 | 0.6599 | 0.182 | 0.4924 | 0.1565 | 0.4924 | 0.1524 | 0.4092 | 0.1931 | 0.4862 |
12.788 | 12.0 | 1284 | 12.2541 | 0.2428 | 0.44 | 0.2248 | 0.1213 | 0.202 | 0.5041 | 0.2484 | 0.4125 | 0.4938 | 0.3264 | 0.4726 | 0.7146 | 0.4213 | 0.6428 | 0.2319 | 0.4861 | 0.1937 | 0.4844 | 0.1301 | 0.32 | 0.2371 | 0.536 |
12.788 | 13.0 | 1391 | 13.4175 | 0.1572 | 0.2948 | 0.1408 | 0.0731 | 0.1404 | 0.4109 | 0.208 | 0.3659 | 0.4667 | 0.3022 | 0.4576 | 0.6952 | 0.2418 | 0.6063 | 0.1828 | 0.4835 | 0.1402 | 0.4826 | 0.0575 | 0.2754 | 0.1637 | 0.4858 |
12.788 | 14.0 | 1498 | 13.0720 | 0.2018 | 0.371 | 0.1912 | 0.0984 | 0.1543 | 0.4649 | 0.2229 | 0.3936 | 0.4802 | 0.2958 | 0.4639 | 0.715 | 0.3978 | 0.6392 | 0.2001 | 0.4734 | 0.1828 | 0.4884 | 0.077 | 0.3015 | 0.1512 | 0.4987 |
10.7992 | 15.0 | 1605 | 13.1509 | 0.2093 | 0.3882 | 0.1946 | 0.1149 | 0.1733 | 0.4644 | 0.2398 | 0.4148 | 0.4899 | 0.3288 | 0.4689 | 0.696 | 0.3568 | 0.6099 | 0.2329 | 0.5025 | 0.1722 | 0.4902 | 0.1078 | 0.3415 | 0.1767 | 0.5053 |
10.7992 | 16.0 | 1712 | 13.5416 | 0.1865 | 0.3502 | 0.1691 | 0.0943 | 0.1519 | 0.4479 | 0.2219 | 0.3868 | 0.4732 | 0.3295 | 0.4422 | 0.6977 | 0.3045 | 0.5698 | 0.2009 | 0.4949 | 0.1642 | 0.4728 | 0.0954 | 0.3554 | 0.1676 | 0.4729 |
10.7992 | 17.0 | 1819 | 13.8027 | 0.1419 | 0.2558 | 0.134 | 0.0598 | 0.1091 | 0.3792 | 0.2014 | 0.3602 | 0.4583 | 0.2925 | 0.4245 | 0.6809 | 0.2747 | 0.6063 | 0.1398 | 0.4709 | 0.1221 | 0.4504 | 0.0503 | 0.2877 | 0.1223 | 0.476 |
10.7992 | 18.0 | 1926 | 13.1241 | 0.205 | 0.389 | 0.1896 | 0.1006 | 0.1856 | 0.4619 | 0.2273 | 0.3813 | 0.4531 | 0.2848 | 0.4267 | 0.6944 | 0.3898 | 0.5995 | 0.2296 | 0.4848 | 0.1864 | 0.442 | 0.0754 | 0.2892 | 0.1436 | 0.4498 |
9.7939 | 19.0 | 2033 | 13.4709 | 0.2089 | 0.3828 | 0.1894 | 0.1018 | 0.1797 | 0.4627 | 0.2357 | 0.4 | 0.4783 | 0.3083 | 0.4537 | 0.6973 | 0.4201 | 0.6414 | 0.2105 | 0.5101 | 0.1717 | 0.4598 | 0.115 | 0.3215 | 0.1273 | 0.4587 |
9.7939 | 20.0 | 2140 | 13.6381 | 0.1755 | 0.3379 | 0.1459 | 0.0909 | 0.1501 | 0.4149 | 0.2176 | 0.3734 | 0.441 | 0.2658 | 0.4223 | 0.6946 | 0.3267 | 0.5896 | 0.1912 | 0.4861 | 0.1899 | 0.4446 | 0.0553 | 0.24 | 0.1146 | 0.4449 |
9.7939 | 21.0 | 2247 | 13.6187 | 0.1785 | 0.3454 | 0.159 | 0.0906 | 0.158 | 0.41 | 0.2274 | 0.3754 | 0.4539 | 0.2919 | 0.4327 | 0.6825 | 0.322 | 0.582 | 0.1915 | 0.4861 | 0.1896 | 0.4558 | 0.0585 | 0.2877 | 0.1307 | 0.4578 |
9.7939 | 22.0 | 2354 | 13.6789 | 0.1736 | 0.3213 | 0.1547 | 0.0804 | 0.1572 | 0.3997 | 0.2114 | 0.3971 | 0.4766 | 0.3206 | 0.4618 | 0.7021 | 0.3437 | 0.5995 | 0.2031 | 0.5215 | 0.1468 | 0.4647 | 0.0694 | 0.3477 | 0.1053 | 0.4493 |
9.7939 | 23.0 | 2461 | 13.5973 | 0.1853 | 0.3484 | 0.163 | 0.0873 | 0.1661 | 0.4546 | 0.2201 | 0.3794 | 0.4493 | 0.2877 | 0.4317 | 0.6857 | 0.3301 | 0.5847 | 0.2031 | 0.4937 | 0.1784 | 0.4317 | 0.0745 | 0.2938 | 0.1405 | 0.4427 |
9.0938 | 24.0 | 2568 | 13.2147 | 0.2232 | 0.4117 | 0.2114 | 0.1183 | 0.1953 | 0.4989 | 0.2347 | 0.3911 | 0.4575 | 0.3023 | 0.4273 | 0.6881 | 0.3994 | 0.6032 | 0.204 | 0.4734 | 0.2114 | 0.4621 | 0.1332 | 0.3062 | 0.1683 | 0.4427 |
9.0938 | 25.0 | 2675 | 13.4741 | 0.183 | 0.3428 | 0.1701 | 0.0874 | 0.1512 | 0.4309 | 0.2035 | 0.3643 | 0.4371 | 0.2686 | 0.4134 | 0.6784 | 0.3489 | 0.5842 | 0.1923 | 0.4443 | 0.1939 | 0.4469 | 0.0618 | 0.2938 | 0.1183 | 0.4164 |
9.0938 | 26.0 | 2782 | 13.6609 | 0.1955 | 0.3825 | 0.1691 | 0.1002 | 0.1798 | 0.4468 | 0.2164 | 0.3665 | 0.4337 | 0.2749 | 0.4048 | 0.6729 | 0.3661 | 0.5761 | 0.2018 | 0.4392 | 0.1914 | 0.4393 | 0.098 | 0.2954 | 0.1204 | 0.4187 |
9.0938 | 27.0 | 2889 | 13.5787 | 0.2059 | 0.384 | 0.1917 | 0.1021 | 0.1888 | 0.4571 | 0.224 | 0.3871 | 0.4541 | 0.2992 | 0.4273 | 0.6842 | 0.3905 | 0.5806 | 0.2016 | 0.4835 | 0.1938 | 0.45 | 0.1011 | 0.3092 | 0.1423 | 0.4471 |
9.0938 | 28.0 | 2996 | 13.2892 | 0.2043 | 0.3784 | 0.192 | 0.099 | 0.1847 | 0.4577 | 0.2276 | 0.386 | 0.456 | 0.3049 | 0.4245 | 0.6931 | 0.3972 | 0.5991 | 0.1786 | 0.4797 | 0.2112 | 0.4567 | 0.0698 | 0.2938 | 0.1646 | 0.4507 |
8.5991 | 29.0 | 3103 | 13.5481 | 0.2103 | 0.3954 | 0.197 | 0.11 | 0.1973 | 0.4571 | 0.225 | 0.3865 | 0.4592 | 0.3234 | 0.4292 | 0.6822 | 0.3794 | 0.5748 | 0.202 | 0.4823 | 0.2086 | 0.4616 | 0.1024 | 0.3185 | 0.1591 | 0.4591 |
8.5991 | 30.0 | 3210 | 13.5758 | 0.2024 | 0.3781 | 0.186 | 0.1041 | 0.1897 | 0.4491 | 0.2214 | 0.3882 | 0.46 | 0.3184 | 0.4341 | 0.6808 | 0.3815 | 0.5779 | 0.1844 | 0.4734 | 0.2072 | 0.458 | 0.0877 | 0.3323 | 0.1511 | 0.4582 |
Framework versions
- Transformers 4.42.0.dev0
- Pytorch 2.1.0+cu118
- Datasets 2.19.1
- Tokenizers 0.19.1