Super_Detection_Model
This model is a fine-tuned version of facebook/detr-resnet-50 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1666
- Map: 0.1631
- Map 50: 0.3094
- Map 75: 0.1578
- Map Small: 0.136
- Map Medium: 0.2597
- Map Large: 0.1753
- Mar 1: 0.1149
- Mar 10: 0.2246
- Mar 100: 0.2567
- Mar Small: 0.2577
- Mar Medium: 0.3214
- Mar Large: 0.3473
- Map Car: 0.2921
- Mar 100 Car: 0.4101
- Map Hgv: 0.3204
- Mar 100 Hgv: 0.4958
- Map Motorcycle: 0.0399
- Mar 100 Motorcycle: 0.1208
- Map Other: 0.0
- Mar 100 Other: 0.0
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 Car | Mar 100 Car | Map Hgv | Mar 100 Hgv | Map Motorcycle | Mar 100 Motorcycle | Map Other | Mar 100 Other |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2.0243 | 1.0 | 1431 | 1.8261 | 0.0224 | 0.0754 | 0.0071 | 0.0205 | 0.073 | 0.0404 | 0.0153 | 0.0488 | 0.0703 | 0.0623 | 0.1422 | 0.1632 | 0.0879 | 0.2349 | 0.0016 | 0.0462 | 0.0 | 0.0 | 0.0 | 0.0 |
1.8222 | 2.0 | 2862 | 1.7582 | 0.0263 | 0.0793 | 0.011 | 0.0243 | 0.0842 | 0.0618 | 0.0282 | 0.0689 | 0.089 | 0.0681 | 0.1561 | 0.2498 | 0.0964 | 0.2493 | 0.0088 | 0.1068 | 0.0 | 0.0 | 0.0 | 0.0 |
1.7513 | 3.0 | 4293 | 1.8670 | 0.0326 | 0.095 | 0.016 | 0.0262 | 0.0968 | 0.0393 | 0.0394 | 0.0846 | 0.1004 | 0.0673 | 0.1735 | 0.1284 | 0.0925 | 0.2213 | 0.038 | 0.1803 | 0.0 | 0.0 | 0.0 | 0.0 |
1.7318 | 4.0 | 5724 | 1.6184 | 0.0444 | 0.1157 | 0.0264 | 0.0343 | 0.1308 | 0.0531 | 0.054 | 0.1232 | 0.1544 | 0.1096 | 0.2494 | 0.2399 | 0.1154 | 0.2994 | 0.0624 | 0.318 | 0.0 | 0.0 | 0.0 | 0.0 |
1.7005 | 5.0 | 7155 | 1.6096 | 0.0563 | 0.1434 | 0.038 | 0.0381 | 0.1361 | 0.069 | 0.057 | 0.1142 | 0.1376 | 0.0907 | 0.2308 | 0.2079 | 0.1249 | 0.2756 | 0.1002 | 0.2749 | 0.0 | 0.0 | 0.0 | 0.0 |
1.6242 | 6.0 | 8586 | 1.5605 | 0.0566 | 0.1396 | 0.0399 | 0.0408 | 0.1475 | 0.0656 | 0.0599 | 0.1303 | 0.1603 | 0.1173 | 0.2522 | 0.2842 | 0.116 | 0.2961 | 0.1095 | 0.3421 | 0.0008 | 0.003 | 0.0 | 0.0 |
1.6339 | 7.0 | 10017 | 1.5466 | 0.0677 | 0.1558 | 0.0514 | 0.0509 | 0.1477 | 0.0708 | 0.0609 | 0.1372 | 0.1654 | 0.141 | 0.2481 | 0.2283 | 0.1688 | 0.3031 | 0.1015 | 0.3348 | 0.0007 | 0.0238 | 0.0 | 0.0 |
1.611 | 8.0 | 11448 | 1.6296 | 0.0656 | 0.1601 | 0.0465 | 0.0447 | 0.1513 | 0.0639 | 0.0581 | 0.1229 | 0.1461 | 0.1028 | 0.2445 | 0.2403 | 0.1505 | 0.2771 | 0.1093 | 0.2967 | 0.0025 | 0.0107 | 0.0 | 0.0 |
1.5769 | 9.0 | 12879 | 1.5218 | 0.0817 | 0.1811 | 0.0655 | 0.0614 | 0.1676 | 0.0751 | 0.0699 | 0.1476 | 0.1762 | 0.1485 | 0.2602 | 0.2574 | 0.1837 | 0.2981 | 0.1349 | 0.343 | 0.0083 | 0.0637 | 0.0 | 0.0 |
1.5192 | 10.0 | 14310 | 1.5124 | 0.0858 | 0.1911 | 0.0722 | 0.0558 | 0.1774 | 0.0845 | 0.0726 | 0.1462 | 0.1731 | 0.1398 | 0.2583 | 0.2369 | 0.1615 | 0.3079 | 0.1738 | 0.3261 | 0.0079 | 0.0583 | 0.0 | 0.0 |
1.5417 | 11.0 | 15741 | 1.4493 | 0.0985 | 0.212 | 0.084 | 0.0715 | 0.1935 | 0.0993 | 0.077 | 0.1521 | 0.1775 | 0.1501 | 0.2581 | 0.2797 | 0.1938 | 0.3146 | 0.187 | 0.331 | 0.0133 | 0.0643 | 0.0 | 0.0 |
1.4639 | 12.0 | 17172 | 1.4144 | 0.1015 | 0.2209 | 0.0849 | 0.0756 | 0.1908 | 0.0964 | 0.0779 | 0.1578 | 0.1821 | 0.1632 | 0.2614 | 0.3029 | 0.2114 | 0.3273 | 0.1803 | 0.3355 | 0.0141 | 0.0655 | 0.0 | 0.0 |
1.4552 | 13.0 | 18603 | 1.4097 | 0.098 | 0.2143 | 0.0784 | 0.0756 | 0.1834 | 0.0974 | 0.0793 | 0.172 | 0.2072 | 0.1964 | 0.2771 | 0.2915 | 0.2084 | 0.3444 | 0.1679 | 0.3982 | 0.0159 | 0.0863 | 0.0 | 0.0 |
1.4135 | 14.0 | 20034 | 1.3709 | 0.1232 | 0.2537 | 0.1064 | 0.1001 | 0.2153 | 0.098 | 0.0957 | 0.1857 | 0.2135 | 0.1966 | 0.2919 | 0.3075 | 0.233 | 0.3451 | 0.2224 | 0.4144 | 0.0373 | 0.0946 | 0.0 | 0.0 |
1.3648 | 15.0 | 21465 | 1.3388 | 0.1261 | 0.2621 | 0.1064 | 0.0997 | 0.22 | 0.1053 | 0.0956 | 0.1931 | 0.223 | 0.2157 | 0.2928 | 0.2985 | 0.2378 | 0.3632 | 0.2348 | 0.4318 | 0.0317 | 0.097 | 0.0 | 0.0 |
1.3848 | 16.0 | 22896 | 1.3085 | 0.1221 | 0.2473 | 0.1119 | 0.0891 | 0.2217 | 0.1163 | 0.098 | 0.1889 | 0.2181 | 0.202 | 0.3679 | 0.3138 | 0.223 | 0.3537 | 0.2482 | 0.4275 | 0.0172 | 0.0911 | 0.0 | 0.0 |
1.3514 | 17.0 | 24327 | 1.2929 | 0.1308 | 0.2737 | 0.1113 | 0.0987 | 0.2308 | 0.1232 | 0.1009 | 0.1995 | 0.2286 | 0.2175 | 0.2992 | 0.3208 | 0.2345 | 0.3652 | 0.2544 | 0.4416 | 0.0344 | 0.1077 | 0.0 | 0.0 |
1.3502 | 18.0 | 25758 | 1.2871 | 0.1304 | 0.272 | 0.1161 | 0.0992 | 0.2295 | 0.1483 | 0.0975 | 0.1936 | 0.2225 | 0.2081 | 0.2928 | 0.3418 | 0.2451 | 0.3704 | 0.2546 | 0.4224 | 0.0217 | 0.097 | 0.0 | 0.0 |
1.2679 | 19.0 | 27189 | 1.2635 | 0.1347 | 0.2762 | 0.121 | 0.1035 | 0.2416 | 0.1493 | 0.1051 | 0.2019 | 0.2318 | 0.2171 | 0.4472 | 0.3478 | 0.2477 | 0.3708 | 0.2594 | 0.4334 | 0.0316 | 0.1232 | 0.0 | 0.0 |
1.2864 | 20.0 | 28620 | 1.2418 | 0.1428 | 0.2839 | 0.1278 | 0.1093 | 0.2477 | 0.1554 | 0.1082 | 0.2071 | 0.2397 | 0.2309 | 0.4281 | 0.3248 | 0.2608 | 0.3873 | 0.2756 | 0.4506 | 0.0349 | 0.1208 | 0.0 | 0.0 |
1.2492 | 21.0 | 30051 | 1.2324 | 0.151 | 0.2957 | 0.14 | 0.1187 | 0.2491 | 0.1583 | 0.1131 | 0.2166 | 0.2485 | 0.2471 | 0.314 | 0.329 | 0.2675 | 0.3891 | 0.2996 | 0.4835 | 0.037 | 0.1214 | 0.0 | 0.0 |
1.2282 | 22.0 | 31482 | 1.2092 | 0.1561 | 0.3021 | 0.1477 | 0.1281 | 0.2522 | 0.1707 | 0.1146 | 0.2187 | 0.2519 | 0.2494 | 0.3176 | 0.3488 | 0.2763 | 0.3983 | 0.3116 | 0.4877 | 0.0364 | 0.1214 | 0.0 | 0.0 |
1.2218 | 23.0 | 32913 | 1.1896 | 0.1533 | 0.2986 | 0.1426 | 0.1261 | 0.2506 | 0.1555 | 0.1111 | 0.2177 | 0.25 | 0.25 | 0.3149 | 0.3495 | 0.2836 | 0.4018 | 0.2932 | 0.4759 | 0.0364 | 0.122 | 0.0 | 0.0 |
1.2083 | 24.0 | 34344 | 1.1887 | 0.1536 | 0.2964 | 0.1448 | 0.1262 | 0.2524 | 0.163 | 0.1109 | 0.2198 | 0.2531 | 0.2541 | 0.3929 | 0.3437 | 0.2792 | 0.4027 | 0.3017 | 0.4887 | 0.0334 | 0.1208 | 0.0 | 0.0 |
1.2021 | 25.0 | 35775 | 1.1857 | 0.1584 | 0.3051 | 0.1519 | 0.131 | 0.2559 | 0.1703 | 0.1129 | 0.2198 | 0.2543 | 0.2557 | 0.3201 | 0.3446 | 0.2893 | 0.4064 | 0.3095 | 0.4966 | 0.0349 | 0.1143 | 0.0 | 0.0 |
1.2054 | 26.0 | 37206 | 1.1760 | 0.1609 | 0.3088 | 0.1541 | 0.1345 | 0.2567 | 0.1787 | 0.1159 | 0.2227 | 0.2546 | 0.253 | 0.3211 | 0.349 | 0.2939 | 0.4076 | 0.3108 | 0.4893 | 0.0387 | 0.1214 | 0.0 | 0.0 |
1.1786 | 27.0 | 38637 | 1.1732 | 0.1612 | 0.3063 | 0.1565 | 0.1328 | 0.258 | 0.174 | 0.1141 | 0.2242 | 0.2564 | 0.2532 | 0.3232 | 0.3475 | 0.292 | 0.4087 | 0.3157 | 0.4973 | 0.0372 | 0.1196 | 0.0 | 0.0 |
1.1786 | 28.0 | 40068 | 1.1681 | 0.1635 | 0.3102 | 0.1562 | 0.1372 | 0.2602 | 0.1769 | 0.1157 | 0.2248 | 0.2573 | 0.2583 | 0.3225 | 0.3481 | 0.2919 | 0.4104 | 0.3222 | 0.499 | 0.0399 | 0.1196 | 0.0 | 0.0 |
1.1715 | 29.0 | 41499 | 1.1684 | 0.1628 | 0.3095 | 0.1571 | 0.1359 | 0.2592 | 0.1755 | 0.1149 | 0.2242 | 0.2563 | 0.257 | 0.321 | 0.3474 | 0.2909 | 0.409 | 0.3196 | 0.4954 | 0.0407 | 0.1208 | 0.0 | 0.0 |
1.1823 | 30.0 | 42930 | 1.1666 | 0.1631 | 0.3094 | 0.1578 | 0.136 | 0.2597 | 0.1753 | 0.1149 | 0.2246 | 0.2567 | 0.2577 | 0.3214 | 0.3473 | 0.2921 | 0.4101 | 0.3204 | 0.4958 | 0.0399 | 0.1208 | 0.0 | 0.0 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.4.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 54