<!-- 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. -->
doc-segment
This model is a fine-tuned version of nvidia/mit-b0 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0665
- Mean Iou: 0.7633
- Mean Accuracy: 1.0
- Overall Accuracy: 1.0
- Per Category Iou: [0.7632902145385743]
- Per Category Accuracy: [1.0]
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: 6e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Per Category Iou | Per Category Accuracy |
|---|---|---|---|---|---|---|---|---|
| 0.2461 | 0.38 | 15 | 0.3998 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.3589 | 0.75 | 30 | 0.3173 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.2486 | 1.12 | 45 | 0.3038 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.3872 | 1.5 | 60 | 0.2414 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.2428 | 1.88 | 75 | 0.2138 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.267 | 2.25 | 90 | 0.2384 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1593 | 2.62 | 105 | 0.1965 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1945 | 3.0 | 120 | 0.1901 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1439 | 3.38 | 135 | 0.1763 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1778 | 3.75 | 150 | 0.1817 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1387 | 4.12 | 165 | 0.1603 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1361 | 4.5 | 180 | 0.1420 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1231 | 4.88 | 195 | 0.1482 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0922 | 5.25 | 210 | 0.1338 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1217 | 5.62 | 225 | 0.1408 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1138 | 6.0 | 240 | 0.1352 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0944 | 6.38 | 255 | 0.1266 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1211 | 6.75 | 270 | 0.1249 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1213 | 7.12 | 285 | 0.1158 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0707 | 7.5 | 300 | 0.1192 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0869 | 7.88 | 315 | 0.1146 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1362 | 8.25 | 330 | 0.1101 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0748 | 8.62 | 345 | 0.1028 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0684 | 9.0 | 360 | 0.0876 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0686 | 9.38 | 375 | 0.0922 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0773 | 9.75 | 390 | 0.1011 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0694 | 10.12 | 405 | 0.0955 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0588 | 10.5 | 420 | 0.0912 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1351 | 10.88 | 435 | 0.1102 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0719 | 11.25 | 450 | 0.0926 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0891 | 11.62 | 465 | 0.0895 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.048 | 12.0 | 480 | 0.0900 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0642 | 12.38 | 495 | 0.0853 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1318 | 12.75 | 510 | 0.0877 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0528 | 13.12 | 525 | 0.0820 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.071 | 13.5 | 540 | 0.0885 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0938 | 13.88 | 555 | 0.0873 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0456 | 14.25 | 570 | 0.0760 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0432 | 14.62 | 585 | 0.0750 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0492 | 15.0 | 600 | 0.0751 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0469 | 15.38 | 615 | 0.0689 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0508 | 15.75 | 630 | 0.0765 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0523 | 16.12 | 645 | 0.0766 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.1041 | 16.5 | 660 | 0.0758 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0489 | 16.88 | 675 | 0.0734 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.047 | 17.25 | 690 | 0.0718 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0582 | 17.62 | 705 | 0.0788 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0385 | 18.0 | 720 | 0.0726 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0328 | 18.38 | 735 | 0.0689 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0464 | 18.75 | 750 | 0.0748 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0532 | 19.12 | 765 | 0.0762 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0532 | 19.5 | 780 | 0.0757 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0522 | 19.88 | 795 | 0.0745 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0375 | 20.25 | 810 | 0.0732 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0392 | 20.62 | 825 | 0.0670 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0391 | 21.0 | 840 | 0.0702 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0379 | 21.38 | 855 | 0.0658 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.094 | 21.75 | 870 | 0.0725 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.038 | 22.12 | 885 | 0.0676 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0681 | 22.5 | 900 | 0.0734 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0344 | 22.88 | 915 | 0.0653 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0363 | 23.25 | 930 | 0.0613 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0451 | 23.62 | 945 | 0.0716 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0539 | 24.0 | 960 | 0.0708 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.075 | 24.38 | 975 | 0.0781 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0427 | 24.75 | 990 | 0.0659 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0356 | 25.12 | 1005 | 0.0711 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0466 | 25.5 | 1020 | 0.0652 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0349 | 25.88 | 1035 | 0.0632 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0288 | 26.25 | 1050 | 0.0650 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0361 | 26.62 | 1065 | 0.0656 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0463 | 27.0 | 1080 | 0.0632 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0426 | 27.38 | 1095 | 0.0666 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0323 | 27.75 | 1110 | 0.0651 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0394 | 28.12 | 1125 | 0.0643 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0673 | 28.5 | 1140 | 0.0657 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0725 | 28.88 | 1155 | 0.0675 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0361 | 29.25 | 1170 | 0.0654 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0269 | 29.62 | 1185 | 0.0617 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
| 0.0494 | 30.0 | 1200 | 0.0665 | 0.7633 | 1.0 | 1.0 | [0.7632902145385743] | [1.0] |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
