metadata
library_name: transformers
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 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.6381
- Map: 0.1544
- Map 50: 0.3211
- Map 75: 0.1335
- Map Small: 0.0306
- Map Medium: 0.1213
- Map Large: 0.2305
- Mar 1: 0.1679
- Mar 10: 0.3519
- Mar 100: 0.391
- Mar Small: 0.1517
- Mar Medium: 0.3448
- Mar Large: 0.5355
- Map Coverall: 0.453
- Mar 100 Coverall: 0.6428
- Map Face Shield: 0.0221
- Mar 100 Face Shield: 0.3165
- Map Gloves: 0.0499
- Mar 100 Gloves: 0.342
- Map Goggles: 0.0449
- Mar 100 Goggles: 0.2846
- Map Mask: 0.202
- Mar 100 Mask: 0.3693
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: 48
- 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 | 18 | 6.9596 | 0.0001 | 0.0005 | 0.0 | 0.0 | 0.0 | 0.0002 | 0.002 | 0.005 | 0.0144 | 0.0 | 0.0095 | 0.0214 | 0.0003 | 0.0599 | 0.0 | 0.0 | 0.0 | 0.0013 | 0.0 | 0.0 | 0.0001 | 0.0107 |
No log | 2.0 | 36 | 3.3034 | 0.0011 | 0.0045 | 0.0003 | 0.0011 | 0.0014 | 0.0031 | 0.0048 | 0.0328 | 0.0561 | 0.032 | 0.0559 | 0.0517 | 0.0021 | 0.1086 | 0.0 | 0.0063 | 0.0002 | 0.0263 | 0.0017 | 0.0508 | 0.0014 | 0.0884 |
No log | 3.0 | 54 | 2.8530 | 0.0039 | 0.0115 | 0.002 | 0.0008 | 0.0014 | 0.0113 | 0.0167 | 0.0698 | 0.1033 | 0.0214 | 0.0651 | 0.1309 | 0.0156 | 0.3122 | 0.0 | 0.0101 | 0.0011 | 0.0719 | 0.0002 | 0.0308 | 0.0025 | 0.0916 |
No log | 4.0 | 72 | 2.5424 | 0.0156 | 0.0329 | 0.0151 | 0.0015 | 0.0092 | 0.0199 | 0.0413 | 0.0949 | 0.1383 | 0.0522 | 0.1064 | 0.1807 | 0.0689 | 0.3369 | 0.0001 | 0.0013 | 0.002 | 0.1321 | 0.0005 | 0.0385 | 0.0063 | 0.1827 |
No log | 5.0 | 90 | 2.4114 | 0.0201 | 0.0454 | 0.0151 | 0.0046 | 0.0166 | 0.0272 | 0.0528 | 0.1104 | 0.1562 | 0.0615 | 0.1094 | 0.2159 | 0.079 | 0.3577 | 0.0 | 0.0 | 0.0043 | 0.1723 | 0.0 | 0.0 | 0.0172 | 0.2511 |
No log | 6.0 | 108 | 2.2955 | 0.0352 | 0.0765 | 0.0298 | 0.0072 | 0.0221 | 0.0371 | 0.067 | 0.1331 | 0.1811 | 0.0667 | 0.1508 | 0.2155 | 0.1438 | 0.4477 | 0.0 | 0.0 | 0.0052 | 0.1737 | 0.0052 | 0.0169 | 0.0218 | 0.2671 |
No log | 7.0 | 126 | 2.2319 | 0.0333 | 0.0741 | 0.0268 | 0.0098 | 0.0259 | 0.0484 | 0.0829 | 0.1645 | 0.2172 | 0.0771 | 0.1655 | 0.2879 | 0.1201 | 0.491 | 0.0 | 0.0 | 0.0103 | 0.2362 | 0.0018 | 0.0585 | 0.0343 | 0.3004 |
No log | 8.0 | 144 | 2.1602 | 0.0346 | 0.0762 | 0.027 | 0.0105 | 0.0319 | 0.0475 | 0.0822 | 0.1698 | 0.2212 | 0.0784 | 0.166 | 0.2791 | 0.114 | 0.5378 | 0.0012 | 0.0063 | 0.0061 | 0.2027 | 0.0036 | 0.0385 | 0.0482 | 0.3204 |
No log | 9.0 | 162 | 2.1318 | 0.0365 | 0.0751 | 0.0317 | 0.008 | 0.0388 | 0.0493 | 0.0876 | 0.1838 | 0.2341 | 0.0799 | 0.1945 | 0.2747 | 0.115 | 0.5775 | 0.0 | 0.0 | 0.0058 | 0.1929 | 0.0054 | 0.0646 | 0.0562 | 0.3356 |
No log | 10.0 | 180 | 2.0494 | 0.0454 | 0.1034 | 0.0363 | 0.0132 | 0.0438 | 0.0645 | 0.1057 | 0.2013 | 0.2497 | 0.0892 | 0.1899 | 0.36 | 0.1363 | 0.5279 | 0.0001 | 0.0051 | 0.0082 | 0.2491 | 0.0086 | 0.0923 | 0.0736 | 0.3742 |
No log | 11.0 | 198 | 2.0013 | 0.0505 | 0.1115 | 0.0411 | 0.0108 | 0.0482 | 0.0709 | 0.1015 | 0.2235 | 0.269 | 0.0854 | 0.2097 | 0.3776 | 0.1646 | 0.5914 | 0.0005 | 0.0177 | 0.0113 | 0.2562 | 0.0061 | 0.1215 | 0.0699 | 0.3582 |
No log | 12.0 | 216 | 1.9699 | 0.057 | 0.1211 | 0.0445 | 0.0117 | 0.0476 | 0.078 | 0.0962 | 0.2212 | 0.2676 | 0.0758 | 0.215 | 0.3634 | 0.1992 | 0.6122 | 0.0004 | 0.0127 | 0.0093 | 0.2527 | 0.006 | 0.1169 | 0.0702 | 0.3436 |
No log | 13.0 | 234 | 1.9105 | 0.0722 | 0.1588 | 0.06 | 0.0183 | 0.0591 | 0.1058 | 0.1318 | 0.2622 | 0.3075 | 0.1026 | 0.2644 | 0.4241 | 0.2304 | 0.6252 | 0.0018 | 0.062 | 0.0125 | 0.2848 | 0.0221 | 0.1862 | 0.0942 | 0.3791 |
No log | 14.0 | 252 | 1.8849 | 0.0859 | 0.1809 | 0.0771 | 0.0189 | 0.0681 | 0.1184 | 0.1271 | 0.2626 | 0.3093 | 0.1056 | 0.265 | 0.4173 | 0.2761 | 0.632 | 0.0044 | 0.0962 | 0.0153 | 0.2835 | 0.0203 | 0.1615 | 0.1136 | 0.3733 |
No log | 15.0 | 270 | 1.8380 | 0.0968 | 0.2026 | 0.0867 | 0.0139 | 0.0679 | 0.1375 | 0.1275 | 0.2733 | 0.3172 | 0.1078 | 0.2588 | 0.4298 | 0.3325 | 0.645 | 0.0111 | 0.1367 | 0.0173 | 0.3022 | 0.0124 | 0.1369 | 0.1108 | 0.3653 |
No log | 16.0 | 288 | 1.8123 | 0.1153 | 0.2438 | 0.101 | 0.0254 | 0.0862 | 0.1513 | 0.14 | 0.2974 | 0.3346 | 0.1297 | 0.2825 | 0.443 | 0.3832 | 0.6392 | 0.0256 | 0.2114 | 0.0192 | 0.3125 | 0.0221 | 0.16 | 0.1265 | 0.3498 |
No log | 17.0 | 306 | 1.7964 | 0.1199 | 0.2621 | 0.1026 | 0.0219 | 0.094 | 0.1591 | 0.1306 | 0.2957 | 0.3384 | 0.1264 | 0.2885 | 0.454 | 0.3926 | 0.6374 | 0.0236 | 0.1987 | 0.0221 | 0.3152 | 0.0271 | 0.2062 | 0.1343 | 0.3347 |
No log | 18.0 | 324 | 1.7520 | 0.1294 | 0.2814 | 0.1075 | 0.0242 | 0.1067 | 0.1774 | 0.1399 | 0.319 | 0.3555 | 0.138 | 0.3067 | 0.4831 | 0.4066 | 0.6541 | 0.0262 | 0.2165 | 0.0302 | 0.3263 | 0.0286 | 0.2246 | 0.1555 | 0.356 |
No log | 19.0 | 342 | 1.7232 | 0.1373 | 0.2907 | 0.1166 | 0.0273 | 0.1082 | 0.1956 | 0.1483 | 0.3258 | 0.3608 | 0.1471 | 0.3052 | 0.5008 | 0.4284 | 0.645 | 0.016 | 0.2266 | 0.0355 | 0.3402 | 0.0305 | 0.2292 | 0.1764 | 0.3631 |
No log | 20.0 | 360 | 1.7113 | 0.1395 | 0.3024 | 0.1141 | 0.0301 | 0.1092 | 0.2093 | 0.1525 | 0.3247 | 0.3575 | 0.1575 | 0.3006 | 0.4969 | 0.4258 | 0.6293 | 0.0225 | 0.2418 | 0.0372 | 0.3366 | 0.0254 | 0.2277 | 0.1869 | 0.352 |
No log | 21.0 | 378 | 1.6864 | 0.1447 | 0.3079 | 0.1238 | 0.0295 | 0.1154 | 0.2157 | 0.1598 | 0.342 | 0.374 | 0.1575 | 0.3211 | 0.5222 | 0.4284 | 0.6437 | 0.0231 | 0.2671 | 0.045 | 0.3384 | 0.0349 | 0.2615 | 0.1923 | 0.3591 |
No log | 22.0 | 396 | 1.6746 | 0.1495 | 0.3155 | 0.1282 | 0.03 | 0.115 | 0.2223 | 0.169 | 0.3466 | 0.379 | 0.1673 | 0.3227 | 0.5279 | 0.4376 | 0.6464 | 0.0258 | 0.2759 | 0.0458 | 0.3362 | 0.0436 | 0.2677 | 0.1946 | 0.3689 |
No log | 23.0 | 414 | 1.6604 | 0.1499 | 0.311 | 0.1336 | 0.0313 | 0.1178 | 0.2233 | 0.161 | 0.3479 | 0.3836 | 0.1599 | 0.3352 | 0.5265 | 0.4435 | 0.6486 | 0.0246 | 0.3013 | 0.0458 | 0.3339 | 0.0411 | 0.2677 | 0.1944 | 0.3662 |
No log | 24.0 | 432 | 1.6552 | 0.1508 | 0.3167 | 0.1301 | 0.0284 | 0.1209 | 0.2256 | 0.1645 | 0.3503 | 0.389 | 0.1621 | 0.342 | 0.533 | 0.4469 | 0.6446 | 0.0229 | 0.3025 | 0.046 | 0.3429 | 0.0399 | 0.2846 | 0.1985 | 0.3702 |
No log | 25.0 | 450 | 1.6465 | 0.1506 | 0.3124 | 0.13 | 0.0287 | 0.1185 | 0.2266 | 0.1611 | 0.3505 | 0.3869 | 0.1588 | 0.339 | 0.5355 | 0.4472 | 0.6446 | 0.0209 | 0.2962 | 0.0473 | 0.342 | 0.0404 | 0.2831 | 0.1974 | 0.3684 |
No log | 26.0 | 468 | 1.6419 | 0.1526 | 0.3209 | 0.1298 | 0.0283 | 0.1197 | 0.2258 | 0.1625 | 0.3453 | 0.3854 | 0.1545 | 0.338 | 0.5263 | 0.4531 | 0.6428 | 0.022 | 0.2911 | 0.0467 | 0.3438 | 0.0441 | 0.2785 | 0.197 | 0.3707 |
No log | 27.0 | 486 | 1.6383 | 0.1546 | 0.3235 | 0.1354 | 0.0288 | 0.1247 | 0.2274 | 0.164 | 0.3475 | 0.3872 | 0.1546 | 0.3396 | 0.5312 | 0.4539 | 0.645 | 0.0216 | 0.2975 | 0.0493 | 0.3438 | 0.048 | 0.2815 | 0.2004 | 0.3684 |
3.1149 | 28.0 | 504 | 1.6393 | 0.1544 | 0.3208 | 0.1328 | 0.0306 | 0.1218 | 0.2298 | 0.1668 | 0.3518 | 0.3902 | 0.1524 | 0.3423 | 0.5368 | 0.4535 | 0.6414 | 0.0222 | 0.3114 | 0.0496 | 0.3429 | 0.045 | 0.2862 | 0.2015 | 0.3693 |
3.1149 | 29.0 | 522 | 1.6383 | 0.1544 | 0.3221 | 0.1332 | 0.0308 | 0.1218 | 0.2306 | 0.167 | 0.3515 | 0.3909 | 0.1517 | 0.3436 | 0.536 | 0.4533 | 0.6428 | 0.0219 | 0.3152 | 0.05 | 0.342 | 0.0449 | 0.2846 | 0.2021 | 0.3698 |
3.1149 | 30.0 | 540 | 1.6381 | 0.1544 | 0.3211 | 0.1335 | 0.0306 | 0.1213 | 0.2305 | 0.1679 | 0.3519 | 0.391 | 0.1517 | 0.3448 | 0.5355 | 0.453 | 0.6428 | 0.0221 | 0.3165 | 0.0499 | 0.342 | 0.0449 | 0.2846 | 0.202 | 0.3693 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1