generated_from_trainer

<!-- 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. -->

segformer-b5-finetuned-segments-crop_crack_early-lr6-350-8

This model is a fine-tuned version of nvidia/mit-b5 on the None dataset. It achieves the following results on the evaluation set:

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:

Training results

Training Loss Epoch Step Validation Loss Mean Iou Mean Accuracy Overall Accuracy Accuracy Unlabeled Accuracy Crack Accuracy Potholes Iou Unlabeled Iou Crack Iou Potholes
0.3866 0.07 20 0.4031 0.0249 0.0377 0.0385 nan 0.0424 0.0331 0.0 0.0421 0.0327
0.1963 0.14 40 0.2341 0.0897 0.1376 0.1429 nan 0.1684 0.1068 0.0 0.1624 0.1066
0.1268 0.21 60 0.1775 0.1088 0.1700 0.1885 nan 0.2767 0.0633 0.0 0.2632 0.0633
0.0964 0.28 80 0.1500 0.2160 0.3305 0.3313 nan 0.3349 0.3261 0.0 0.3251 0.3229
0.2564 0.35 100 0.1549 0.1962 0.3067 0.3214 nan 0.3914 0.2221 0.0 0.3666 0.2219
0.1414 0.42 120 0.1507 0.3000 0.4611 0.4689 nan 0.5057 0.4165 0.0 0.4954 0.4045
0.1419 0.49 140 0.1390 0.2567 0.4095 0.3786 nan 0.2310 0.5880 0.0 0.2298 0.5403
0.1134 0.56 160 0.1325 0.2672 0.4087 0.4090 nan 0.4106 0.4069 0.0 0.3988 0.4027
0.1312 0.63 180 0.1262 0.3163 0.4822 0.4740 nan 0.4349 0.5295 0.0 0.4230 0.5259
0.1103 0.7 200 0.1376 0.1918 0.2951 0.2923 nan 0.2791 0.3110 0.0 0.2763 0.2989
0.1163 0.77 220 0.1334 0.2892 0.4453 0.4505 nan 0.4751 0.4156 0.0 0.4523 0.4153
0.1068 0.84 240 0.1237 0.2900 0.4439 0.4374 nan 0.4063 0.4814 0.0 0.3947 0.4753
0.2267 0.91 260 0.1322 0.3473 0.5482 0.5329 nan 0.4600 0.6363 0.0 0.4544 0.5875
0.0842 0.98 280 0.1205 0.3034 0.4680 0.4395 nan 0.3034 0.6326 0.0 0.2989 0.6114
0.1425 1.05 300 0.1226 0.3565 0.5556 0.5476 nan 0.5093 0.6019 0.0 0.5023 0.5674
0.0876 1.12 320 0.1088 0.3571 0.5432 0.5247 nan 0.4364 0.6500 0.0 0.4280 0.6432
0.0829 1.19 340 0.1187 0.2714 0.4112 0.4062 nan 0.3819 0.4405 0.0 0.3761 0.4382
0.11 1.26 360 0.1169 0.3105 0.4726 0.4702 nan 0.4589 0.4863 0.0 0.4505 0.4811
0.1219 1.33 380 0.1144 0.3198 0.4932 0.4804 nan 0.4191 0.5674 0.0 0.4160 0.5435
0.1124 1.4 400 0.1099 0.2951 0.4479 0.4354 nan 0.3758 0.5199 0.0 0.3699 0.5154
0.0729 1.47 420 0.1063 0.3745 0.5697 0.5436 nan 0.4191 0.7203 0.0 0.4124 0.7110
0.1154 1.54 440 0.1079 0.3125 0.4810 0.4525 nan 0.3165 0.6455 0.0 0.3137 0.6238
0.1261 1.61 460 0.1114 0.3406 0.5243 0.5091 nan 0.4368 0.6117 0.0 0.4284 0.5932
0.1003 1.68 480 0.1178 0.2676 0.4071 0.3937 nan 0.3300 0.4841 0.0 0.3190 0.4839
0.0915 1.75 500 0.1035 0.3572 0.5527 0.5296 nan 0.4197 0.6857 0.0 0.4166 0.6549
0.0883 1.82 520 0.1084 0.3561 0.5422 0.5307 nan 0.4761 0.6083 0.0 0.4651 0.6033
0.0766 1.89 540 0.1062 0.3534 0.5374 0.5258 nan 0.4704 0.6045 0.0 0.4586 0.6016
0.1108 1.96 560 0.1169 0.2802 0.4267 0.4250 nan 0.4166 0.4368 0.0 0.4047 0.4359
0.1647 2.03 580 0.1126 0.2839 0.4346 0.4198 nan 0.3493 0.5199 0.0 0.3466 0.5052
0.0811 2.1 600 0.1055 0.3803 0.5800 0.5670 nan 0.5049 0.6551 0.0 0.4972 0.6436
0.0809 2.17 620 0.1157 0.3494 0.5339 0.5334 nan 0.5312 0.5365 0.0 0.5222 0.5261
0.1387 2.24 640 0.1167 0.2710 0.4118 0.3985 nan 0.3352 0.4884 0.0 0.3322 0.4807
0.1004 2.31 660 0.1048 0.3148 0.4783 0.4523 nan 0.3282 0.6285 0.0 0.3253 0.6190
0.0807 2.38 680 0.1032 0.3410 0.5201 0.4996 nan 0.4020 0.6381 0.0 0.3971 0.6259
0.0494 2.45 700 0.1098 0.3095 0.4701 0.4515 nan 0.3629 0.5773 0.0 0.3567 0.5719
0.1155 2.52 720 0.1125 0.3477 0.5306 0.5182 nan 0.4593 0.6018 0.0 0.4465 0.5966
0.0935 2.59 740 0.1141 0.3113 0.4756 0.4651 nan 0.4150 0.5362 0.0 0.4060 0.5278
0.0757 2.66 760 0.1029 0.3753 0.5740 0.5574 nan 0.4780 0.6701 0.0 0.4717 0.6542
0.0705 2.73 780 0.1052 0.3040 0.4633 0.4343 nan 0.2960 0.6306 0.0 0.2945 0.6173
0.0712 2.8 800 0.1069 0.3359 0.5107 0.5001 nan 0.4495 0.5719 0.0 0.4379 0.5697
0.1333 2.87 820 0.1026 0.3495 0.5316 0.5068 nan 0.3885 0.6746 0.0 0.3841 0.6643
0.1269 2.94 840 0.1371 0.2208 0.3346 0.3209 nan 0.2555 0.4137 0.0 0.2489 0.4135
0.1034 3.01 860 0.1012 0.3639 0.5571 0.5373 nan 0.4427 0.6715 0.0 0.4375 0.6541
0.112 3.08 880 0.1007 0.3550 0.5445 0.5165 nan 0.3829 0.7061 0.0 0.3812 0.6838
0.1031 3.15 900 0.1030 0.3416 0.5183 0.4942 nan 0.3794 0.6572 0.0 0.3732 0.6517
0.1174 3.22 920 0.1041 0.3561 0.5413 0.5249 nan 0.4469 0.6357 0.0 0.4392 0.6291
0.1096 3.29 940 0.1132 0.3231 0.4887 0.4733 nan 0.3995 0.5779 0.0 0.3936 0.5756
0.1089 3.36 960 0.1065 0.3735 0.5676 0.5456 nan 0.4406 0.6946 0.0 0.4323 0.6882
0.1207 3.43 980 0.1005 0.3610 0.5493 0.5254 nan 0.4116 0.6869 0.0 0.4067 0.6764
0.0988 3.5 1000 0.1075 0.3235 0.4943 0.4702 nan 0.3550 0.6336 0.0 0.3524 0.6181
0.1012 3.57 1020 0.1108 0.3094 0.4702 0.4488 nan 0.3468 0.5936 0.0 0.3438 0.5844
0.1522 3.64 1040 0.1074 0.3352 0.5083 0.4894 nan 0.3991 0.6175 0.0 0.3929 0.6126
0.0507 3.71 1060 0.1031 0.3508 0.5334 0.5140 nan 0.4214 0.6455 0.0 0.4152 0.6374
0.0974 3.78 1080 0.1108 0.3120 0.4726 0.4498 nan 0.3409 0.6043 0.0 0.3337 0.6024
0.0432 3.85 1100 0.1029 0.3276 0.4984 0.4708 nan 0.3391 0.6577 0.0 0.3346 0.6482
0.0878 3.92 1120 0.1203 0.2935 0.4450 0.4284 nan 0.3489 0.5410 0.0 0.3410 0.5396
0.0645 3.99 1140 0.1114 0.3547 0.5413 0.5229 nan 0.4352 0.6473 0.0 0.4275 0.6365
0.073 4.06 1160 0.1088 0.3599 0.5455 0.5284 nan 0.4466 0.6445 0.0 0.4388 0.6408
0.1115 4.13 1180 0.1039 0.3405 0.5211 0.4964 nan 0.3784 0.6638 0.0 0.3748 0.6468
0.0932 4.2 1200 0.1090 0.3421 0.5193 0.5024 nan 0.4219 0.6167 0.0 0.4123 0.6141
0.0625 4.27 1220 0.1057 0.3567 0.5440 0.5180 nan 0.3941 0.6939 0.0 0.3869 0.6832
0.092 4.34 1240 0.1145 0.3060 0.4637 0.4485 nan 0.3761 0.5512 0.0 0.3687 0.5492
0.0605 4.41 1260 0.0981 0.3715 0.5661 0.5389 nan 0.4089 0.7232 0.0 0.4032 0.7113
0.1164 4.48 1280 0.1272 0.3301 0.5069 0.5158 nan 0.5586 0.4551 0.0 0.5355 0.4547
0.1143 4.55 1300 0.1199 0.3206 0.4898 0.4868 nan 0.4726 0.5071 0.0 0.4619 0.4999
0.0704 4.62 1320 0.1187 0.3217 0.4909 0.4857 nan 0.4609 0.5210 0.0 0.4449 0.5202
0.0756 4.69 1340 0.1096 0.3272 0.4973 0.4868 nan 0.4368 0.5578 0.0 0.4274 0.5544
0.0925 4.76 1360 0.1049 0.3392 0.5157 0.4883 nan 0.3577 0.6736 0.0 0.3529 0.6646
0.091 4.83 1380 0.1020 0.3688 0.5637 0.5415 nan 0.4355 0.6920 0.0 0.4294 0.6769
0.0721 4.9 1400 0.1086 0.3267 0.4972 0.4690 nan 0.3342 0.6602 0.0 0.3302 0.6500
0.065 4.97 1420 0.1184 0.3388 0.5174 0.5071 nan 0.4581 0.5767 0.0 0.4442 0.5723
0.0982 5.03 1440 0.1124 0.3311 0.5042 0.4801 nan 0.3654 0.6430 0.0 0.3587 0.6347
0.0564 5.1 1460 0.1055 0.3691 0.5663 0.5456 nan 0.4470 0.6855 0.0 0.4386 0.6688
0.0704 5.17 1480 0.1111 0.3261 0.4949 0.4716 nan 0.3605 0.6293 0.0 0.3529 0.6253
0.1318 5.24 1500 0.1102 0.3438 0.5233 0.5123 nan 0.4599 0.5867 0.0 0.4472 0.5842
0.0679 5.31 1520 0.1099 0.3395 0.5171 0.5042 nan 0.4423 0.5920 0.0 0.4304 0.5880
0.0831 5.38 1540 0.1080 0.3761 0.5723 0.5595 nan 0.4983 0.6463 0.0 0.4865 0.6417
0.0823 5.45 1560 0.1017 0.3696 0.5625 0.5362 nan 0.4109 0.7141 0.0 0.4065 0.7023
0.1767 5.52 1580 0.1148 0.3141 0.4770 0.4594 nan 0.3753 0.5786 0.0 0.3666 0.5757
0.0619 5.59 1600 0.1028 0.3645 0.5625 0.5413 nan 0.4401 0.6849 0.0 0.4374 0.6562
0.0953 5.66 1620 0.1057 0.3654 0.5571 0.5394 nan 0.4551 0.6590 0.0 0.4427 0.6536
0.0784 5.73 1640 0.1070 0.3438 0.5230 0.5007 nan 0.3942 0.6517 0.0 0.3864 0.6450
0.0718 5.8 1660 0.1099 0.3749 0.5724 0.5603 nan 0.5025 0.6423 0.0 0.4845 0.6401
0.0947 5.87 1680 0.1047 0.3769 0.5769 0.5516 nan 0.4309 0.7229 0.0 0.4250 0.7056
0.089 5.94 1700 0.1036 0.3603 0.5478 0.5247 nan 0.4146 0.6810 0.0 0.4086 0.6722
0.0513 6.01 1720 0.1097 0.3378 0.5147 0.4878 nan 0.3592 0.6703 0.0 0.3508 0.6626
0.0876 6.08 1740 0.1042 0.3437 0.5247 0.5025 nan 0.3969 0.6524 0.0 0.3894 0.6417
0.0839 6.15 1760 0.1095 0.3352 0.5097 0.4910 nan 0.4016 0.6178 0.0 0.3906 0.6150
0.0465 6.22 1780 0.1077 0.3488 0.5317 0.5106 nan 0.4101 0.6533 0.0 0.4028 0.6437
0.0644 6.29 1800 0.1066 0.3895 0.5951 0.5790 nan 0.5024 0.6877 0.0 0.4961 0.6725
0.0583 6.36 1820 0.1108 0.3356 0.5102 0.4873 nan 0.3776 0.6429 0.0 0.3724 0.6344
0.0589 6.43 1840 0.1112 0.3365 0.5132 0.4847 nan 0.3488 0.6776 0.0 0.3437 0.6659
0.0518 6.5 1860 0.1183 0.3093 0.4700 0.4432 nan 0.3152 0.6247 0.0 0.3121 0.6158
0.0965 6.57 1880 0.1084 0.3805 0.5787 0.5669 nan 0.5101 0.6473 0.0 0.4999 0.6415
0.064 6.64 1900 0.1244 0.3459 0.5278 0.5182 nan 0.4723 0.5832 0.0 0.4568 0.5810
0.0831 6.71 1920 0.1108 0.3398 0.5184 0.4937 nan 0.3758 0.6610 0.0 0.3732 0.6461
0.0686 6.78 1940 0.1194 0.3305 0.5019 0.4972 nan 0.4748 0.5289 0.0 0.4654 0.5262
0.0984 6.85 1960 0.1129 0.3478 0.5284 0.5120 nan 0.4336 0.6232 0.0 0.4241 0.6194
0.074 6.92 1980 0.1088 0.3607 0.5506 0.5286 nan 0.4237 0.6775 0.0 0.4150 0.6671
0.0869 6.99 2000 0.1031 0.3727 0.5668 0.5473 nan 0.4539 0.6798 0.0 0.4444 0.6737
0.0792 7.06 2020 0.0991 0.3862 0.5870 0.5678 nan 0.4761 0.6979 0.0 0.4696 0.6889
0.0606 7.13 2040 0.1031 0.3684 0.5601 0.5351 nan 0.4156 0.7045 0.0 0.4102 0.6950
0.073 7.2 2060 0.1069 0.3700 0.5620 0.5433 nan 0.4540 0.6701 0.0 0.4460 0.6641
0.0598 7.27 2080 0.1148 0.3471 0.5265 0.5150 nan 0.4598 0.5933 0.0 0.4517 0.5897
0.0789 7.34 2100 0.1110 0.3479 0.5341 0.5128 nan 0.4108 0.6574 0.0 0.4055 0.6381
0.0653 7.41 2120 0.1201 0.3077 0.4662 0.4359 nan 0.2910 0.6415 0.0 0.2886 0.6343
0.0682 7.48 2140 0.1061 0.3568 0.5426 0.5170 nan 0.3946 0.6907 0.0 0.3902 0.6802
0.0768 7.55 2160 0.1121 0.3427 0.5196 0.4971 nan 0.3900 0.6492 0.0 0.3843 0.6438
0.0551 7.62 2180 0.1213 0.3079 0.4665 0.4370 nan 0.2965 0.6365 0.0 0.2935 0.6300
0.0698 7.69 2200 0.1057 0.3757 0.5708 0.5441 nan 0.4163 0.7253 0.0 0.4094 0.7176
0.0444 7.76 2220 0.1080 0.3623 0.5489 0.5312 nan 0.4466 0.6512 0.0 0.4398 0.6471
0.0566 7.83 2240 0.1119 0.3388 0.5143 0.4900 nan 0.3737 0.6549 0.0 0.3690 0.6475
0.0581 7.9 2260 0.1193 0.3367 0.5109 0.4963 nan 0.4266 0.5952 0.0 0.4186 0.5915

Framework versions