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. -->

vit-base_rvl-cdip-small_rvl_cdip-NK1000_simkd

This model is a fine-tuned version of WinKawaks/vit-small-patch16-224 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 Accuracy Brier Loss Nll F1 Micro F1 Macro Ece Aurc
0.0787 1.0 1000 0.0770 0.2978 0.9091 2.9717 0.2978 0.2516 0.2150 0.5174
0.0697 2.0 2000 0.0679 0.6372 0.6886 1.8065 0.6372 0.6302 0.4021 0.1388
0.0655 3.0 3000 0.0645 0.7492 0.5738 1.7388 0.7492 0.7460 0.4215 0.0921
0.0628 4.0 4000 0.0631 0.752 0.5394 1.7446 0.752 0.7551 0.3922 0.0837
0.0611 5.0 5000 0.0612 0.768 0.4928 1.5830 0.768 0.7700 0.3710 0.0655
0.0593 6.0 6000 0.0609 0.7598 0.4655 1.5730 0.7598 0.7667 0.3228 0.0802
0.0578 7.0 7000 0.0585 0.8063 0.4195 1.4053 0.8062 0.8065 0.3459 0.0521
0.0566 8.0 8000 0.0581 0.8073 0.3997 1.2957 0.8073 0.8084 0.3207 0.0538
0.0557 9.0 9000 0.0571 0.8287 0.3810 1.3269 0.8287 0.8301 0.3307 0.0473
0.0554 10.0 10000 0.0573 0.8115 0.3780 1.3469 0.8115 0.8128 0.3011 0.0508
0.0546 11.0 11000 0.0563 0.8395 0.3549 1.2882 0.8395 0.8401 0.3197 0.0386
0.0541 12.0 12000 0.0558 0.839 0.3426 1.2653 0.839 0.8401 0.3014 0.0394
0.0536 13.0 13000 0.0553 0.8465 0.3259 1.1941 0.8465 0.8473 0.2980 0.0357
0.0537 14.0 14000 0.0559 0.8303 0.3499 1.2460 0.8303 0.8338 0.2955 0.0427
0.0532 15.0 15000 0.0551 0.8445 0.3296 1.1799 0.8445 0.8453 0.2990 0.0360
0.0529 16.0 16000 0.0549 0.845 0.3224 1.1801 0.845 0.8456 0.2895 0.0364
0.0527 17.0 17000 0.0549 0.849 0.3264 1.1725 0.849 0.8503 0.2991 0.0363
0.0526 18.0 18000 0.0547 0.8518 0.3170 1.1755 0.8518 0.8527 0.2943 0.0334
0.0524 19.0 19000 0.0546 0.8458 0.3213 1.1417 0.8458 0.8466 0.2917 0.0344
0.0522 20.0 20000 0.0544 0.8545 0.3105 1.1512 0.8545 0.8542 0.2891 0.0333
0.052 21.0 21000 0.0542 0.855 0.3120 1.1403 0.855 0.8555 0.2940 0.0333
0.0518 22.0 22000 0.0542 0.854 0.3096 1.1533 0.854 0.8545 0.2893 0.0319
0.0517 23.0 23000 0.0541 0.8545 0.3098 1.1445 0.8545 0.8556 0.2920 0.0315
0.0516 24.0 24000 0.0540 0.8578 0.3097 1.1273 0.8578 0.8586 0.2958 0.0315
0.0514 25.0 25000 0.0540 0.8532 0.3076 1.1579 0.8532 0.8533 0.2849 0.0342
0.0513 26.0 26000 0.0540 0.855 0.3055 1.1269 0.855 0.8563 0.2855 0.0325
0.0511 27.0 27000 0.0538 0.8565 0.3029 1.1571 0.8565 0.8572 0.2827 0.0334
0.051 28.0 28000 0.0538 0.8598 0.3012 1.1409 0.8598 0.8604 0.2851 0.0317
0.0509 29.0 29000 0.0537 0.86 0.3003 1.1525 0.8600 0.8603 0.2839 0.0323
0.0508 30.0 30000 0.0537 0.8575 0.3024 1.1430 0.8575 0.8585 0.2849 0.0319
0.0507 31.0 31000 0.0537 0.8595 0.3015 1.1454 0.8595 0.8603 0.2859 0.0311
0.0507 32.0 32000 0.0537 0.8598 0.3005 1.1463 0.8598 0.8603 0.2847 0.0316
0.0506 33.0 33000 0.0537 0.8598 0.2966 1.1392 0.8598 0.8605 0.2800 0.0309
0.0506 34.0 34000 0.0537 0.8562 0.3018 1.1442 0.8562 0.8574 0.2813 0.0327
0.0505 35.0 35000 0.0537 0.855 0.2995 1.1402 0.855 0.8556 0.2790 0.0324
0.0505 36.0 36000 0.0537 0.8575 0.2980 1.1324 0.8575 0.8582 0.2783 0.0314
0.0504 37.0 37000 0.0538 0.8562 0.2981 1.1429 0.8562 0.8570 0.2770 0.0320
0.0503 38.0 38000 0.0538 0.8565 0.2997 1.1319 0.8565 0.8573 0.2795 0.0324
0.0503 39.0 39000 0.0538 0.857 0.2988 1.1447 0.857 0.8578 0.2791 0.0320
0.0502 40.0 40000 0.0538 0.8588 0.2982 1.1409 0.8588 0.8595 0.2798 0.0320
0.0502 41.0 41000 0.0538 0.8572 0.2982 1.1455 0.8572 0.8580 0.2781 0.0319
0.0502 42.0 42000 0.0538 0.8602 0.2979 1.1357 0.8602 0.8609 0.2809 0.0320
0.0501 43.0 43000 0.0539 0.8568 0.2987 1.1462 0.8568 0.8574 0.2787 0.0322
0.0501 44.0 44000 0.0539 0.8595 0.2974 1.1456 0.8595 0.8602 0.2789 0.0322
0.0501 45.0 45000 0.0539 0.8592 0.2980 1.1460 0.8592 0.8601 0.2792 0.0322
0.05 46.0 46000 0.0539 0.8588 0.2979 1.1441 0.8588 0.8596 0.2787 0.0322
0.05 47.0 47000 0.0540 0.8592 0.2983 1.1501 0.8592 0.8600 0.2793 0.0324
0.05 48.0 48000 0.0540 0.8588 0.2980 1.1462 0.8588 0.8595 0.2787 0.0324
0.05 49.0 49000 0.0540 0.8598 0.2978 1.1507 0.8598 0.8604 0.2793 0.0324
0.05 50.0 50000 0.0540 0.859 0.2977 1.1492 0.859 0.8598 0.2784 0.0325

Framework versions