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. -->

dit-base-small_rvl_cdip-NK1000_hint

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
2.5147 1.0 1000 2.3332 0.659 0.4557 2.1792 0.659 0.6434 0.0525 0.1283
2.0194 2.0 2000 1.9888 0.7225 0.3793 2.1426 0.7225 0.7275 0.0522 0.0910
1.7429 3.0 3000 1.8097 0.759 0.3332 1.9995 0.7590 0.7599 0.0480 0.0713
1.5665 4.0 4000 1.8189 0.7652 0.3417 2.0345 0.7652 0.7676 0.0920 0.0706
1.4286 5.0 5000 1.7489 0.7588 0.3435 2.0233 0.7588 0.7639 0.0654 0.0745
1.3258 6.0 6000 1.7223 0.7837 0.3212 1.9911 0.7837 0.7818 0.0939 0.0595
1.1929 7.0 7000 1.6777 0.8015 0.3090 1.9764 0.8015 0.8045 0.0991 0.0554
1.0707 8.0 8000 1.8240 0.79 0.3365 1.9708 0.79 0.7907 0.1334 0.0588
1.0054 9.0 9000 1.9355 0.7853 0.3482 2.0445 0.7853 0.7861 0.1404 0.0637
0.924 10.0 10000 2.0125 0.799 0.3399 2.0334 0.799 0.7993 0.1471 0.0584
0.9043 11.0 11000 1.9868 0.8075 0.3252 2.0171 0.8075 0.8078 0.1429 0.0526
0.8442 12.0 12000 2.1626 0.8087 0.3339 2.0839 0.8087 0.8075 0.1523 0.0534
0.8158 13.0 13000 2.1123 0.7903 0.3567 2.0193 0.7903 0.7922 0.1592 0.0580
0.8087 14.0 14000 2.1296 0.814 0.3269 2.0417 0.8140 0.8152 0.1482 0.0533
0.7924 15.0 15000 2.0768 0.816 0.3181 2.1244 0.816 0.8152 0.1435 0.0500
0.7574 16.0 16000 2.3326 0.797 0.3596 2.1292 0.797 0.7960 0.1668 0.0590
0.7501 17.0 17000 2.3190 0.8067 0.3475 2.0635 0.8067 0.8106 0.1617 0.0588
0.7097 18.0 18000 2.3543 0.8067 0.3468 2.0987 0.8067 0.8063 0.1636 0.0561
0.6939 19.0 19000 2.3151 0.8067 0.3405 2.0927 0.8067 0.8047 0.1617 0.0546
0.6972 20.0 20000 2.3289 0.8087 0.3409 2.0291 0.8087 0.8113 0.1628 0.0550
0.6869 21.0 21000 2.3878 0.797 0.3567 2.0736 0.797 0.7995 0.1727 0.0599
0.6621 22.0 22000 2.3796 0.8137 0.3323 2.1080 0.8137 0.8138 0.1570 0.0553
0.6496 23.0 23000 2.4264 0.8143 0.3342 2.0650 0.8143 0.8182 0.1598 0.0562
0.6478 24.0 24000 2.3300 0.8167 0.3313 2.0509 0.8167 0.8170 0.1564 0.0552
0.6184 25.0 25000 2.3143 0.8283 0.3178 1.9842 0.8283 0.8284 0.1498 0.0509
0.6203 26.0 26000 2.3665 0.825 0.3181 2.0647 0.825 0.8246 0.1501 0.0530
0.6151 27.0 27000 2.2673 0.8335 0.3042 1.9961 0.8335 0.8351 0.1431 0.0501
0.6054 28.0 28000 2.2927 0.8283 0.3100 2.0229 0.8283 0.8281 0.1490 0.0508
0.5965 29.0 29000 2.4001 0.8223 0.3235 2.0393 0.8223 0.8266 0.1553 0.0536
0.5869 30.0 30000 2.4167 0.821 0.3269 1.9890 0.821 0.8213 0.1576 0.0553
0.5852 31.0 31000 2.4030 0.83 0.3120 2.0384 0.83 0.8304 0.1500 0.0539
0.5741 32.0 32000 2.4800 0.8233 0.3259 2.1183 0.8233 0.8212 0.1561 0.0550
0.5743 33.0 33000 2.4718 0.821 0.3269 2.0593 0.821 0.8191 0.1582 0.0578
0.5694 34.0 34000 2.3638 0.8297 0.3135 2.0746 0.8297 0.8293 0.1484 0.0534
0.5567 35.0 35000 2.3320 0.8313 0.3054 2.0498 0.8313 0.8311 0.1470 0.0524
0.5547 36.0 36000 2.3902 0.8293 0.3138 2.0001 0.8293 0.8294 0.1506 0.0513
0.55 37.0 37000 2.3812 0.822 0.3239 2.0567 0.822 0.8226 0.1560 0.0541
0.5482 38.0 38000 2.3680 0.8333 0.3063 2.0447 0.8333 0.8332 0.1460 0.0496
0.546 39.0 39000 2.3250 0.8323 0.3051 1.9994 0.8323 0.8315 0.1478 0.0504
0.541 40.0 40000 2.3498 0.837 0.3003 2.0228 0.8370 0.8367 0.1456 0.0493
0.5448 41.0 41000 2.3504 0.833 0.3053 1.9875 0.833 0.8323 0.1492 0.0532
0.5365 42.0 42000 2.3421 0.8323 0.3077 2.0985 0.8323 0.8314 0.1477 0.0501
0.5324 43.0 43000 2.2976 0.84 0.2929 1.9862 0.8400 0.8403 0.1418 0.0507
0.5326 44.0 44000 2.3270 0.838 0.2993 2.0043 0.838 0.8384 0.1438 0.0521
0.5303 45.0 45000 2.2919 0.839 0.2957 1.9625 0.839 0.8395 0.1430 0.0494
0.5276 46.0 46000 2.2861 0.8397 0.2973 1.9382 0.8397 0.8402 0.1442 0.0512
0.5279 47.0 47000 2.2930 0.8387 0.2962 1.9738 0.8387 0.8384 0.1456 0.0499
0.5251 48.0 48000 2.2841 0.84 0.2950 1.9888 0.8400 0.8399 0.1441 0.0490
0.5257 49.0 49000 2.2802 0.84 0.2950 1.9827 0.8400 0.8400 0.1443 0.0491
0.5232 50.0 50000 2.2813 0.841 0.2948 1.9789 0.841 0.8409 0.1435 0.0496

Framework versions