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-finetuned-rvlcdip-small_rvl_cdip-NK1000_kd_MSE

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
No log 1.0 167 0.9313 0.5955 0.5896 2.2535 0.5955 0.5760 0.2103 0.1770
No log 2.0 334 0.6941 0.7027 0.4541 1.7696 0.7027 0.6962 0.1756 0.1075
0.9587 3.0 501 0.5806 0.7435 0.3758 1.8114 0.7435 0.7489 0.1225 0.0802
0.9587 4.0 668 0.4847 0.7808 0.3232 1.6093 0.7808 0.7825 0.0897 0.0600
0.9587 5.0 835 0.4888 0.775 0.3291 1.7398 0.775 0.7686 0.0629 0.0681
0.3668 6.0 1002 0.4441 0.789 0.3136 1.5502 0.7890 0.7882 0.0720 0.0626
0.3668 7.0 1169 0.4258 0.7993 0.3009 1.6320 0.7993 0.7994 0.0626 0.0658
0.3668 8.0 1336 0.4813 0.774 0.3395 1.8387 0.774 0.7788 0.0716 0.0811
0.1959 9.0 1503 0.4289 0.7967 0.3111 1.7125 0.7967 0.7985 0.0799 0.0686
0.1959 10.0 1670 0.4380 0.7897 0.3151 1.8001 0.7897 0.7920 0.0802 0.0730
0.1959 11.0 1837 0.4414 0.796 0.3159 1.7450 0.796 0.7956 0.0795 0.0815
0.1249 12.0 2004 0.4507 0.787 0.3243 1.7219 0.787 0.7840 0.0841 0.0767
0.1249 13.0 2171 0.4209 0.802 0.3095 1.7111 0.802 0.8045 0.0825 0.0729
0.1249 14.0 2338 0.4095 0.8007 0.3039 1.5961 0.8007 0.8018 0.0742 0.0743
0.088 15.0 2505 0.4043 0.8125 0.2974 1.6100 0.8125 0.8158 0.0801 0.0740
0.088 16.0 2672 0.4056 0.8083 0.2964 1.6402 0.8083 0.8080 0.0833 0.0681
0.088 17.0 2839 0.4052 0.8103 0.2993 1.6074 0.8103 0.8105 0.0848 0.0780
0.0638 18.0 3006 0.4207 0.8035 0.3066 1.6669 0.8035 0.8075 0.0826 0.0746
0.0638 19.0 3173 0.3981 0.8125 0.2911 1.5687 0.8125 0.8128 0.0836 0.0762
0.0638 20.0 3340 0.3828 0.8207 0.2803 1.5513 0.8207 0.8217 0.0800 0.0627
0.0456 21.0 3507 0.3710 0.821 0.2802 1.4355 0.821 0.8218 0.0913 0.0662
0.0456 22.0 3674 0.3672 0.8247 0.2744 1.4922 0.8247 0.8280 0.0774 0.0615
0.0456 23.0 3841 0.3600 0.8255 0.2727 1.4413 0.8255 0.8256 0.0817 0.0675
0.0289 24.0 4008 0.3650 0.8235 0.2767 1.3874 0.8235 0.8248 0.0818 0.0698
0.0289 25.0 4175 0.3608 0.827 0.2706 1.3223 0.827 0.8279 0.0861 0.0597
0.0289 26.0 4342 0.3572 0.829 0.2687 1.3947 0.8290 0.8300 0.0878 0.0650
0.0176 27.0 4509 0.3516 0.8315 0.2655 1.3000 0.8315 0.8319 0.0866 0.0597
0.0176 28.0 4676 0.3455 0.8337 0.2626 1.3070 0.8337 0.8351 0.0870 0.0602
0.0176 29.0 4843 0.3489 0.8337 0.2656 1.3027 0.8337 0.8347 0.0859 0.0587
0.011 30.0 5010 0.3472 0.8327 0.2639 1.2879 0.8327 0.8336 0.0878 0.0599
0.011 31.0 5177 0.3468 0.8335 0.2642 1.2955 0.8335 0.8341 0.0859 0.0650
0.011 32.0 5344 0.3467 0.8333 0.2635 1.2911 0.8333 0.8341 0.0849 0.0588
0.0076 33.0 5511 0.3430 0.834 0.2601 1.2738 0.834 0.8346 0.0831 0.0609
0.0076 34.0 5678 0.3442 0.8345 0.2626 1.2921 0.8345 0.8353 0.0864 0.0629
0.0076 35.0 5845 0.3431 0.8355 0.2596 1.2790 0.8355 0.8362 0.0860 0.0589
0.0055 36.0 6012 0.3496 0.8297 0.2646 1.2985 0.8297 0.8305 0.0897 0.0642
0.0055 37.0 6179 0.3445 0.8343 0.2605 1.2509 0.8343 0.8348 0.0862 0.0594
0.0055 38.0 6346 0.3473 0.831 0.2628 1.2919 0.831 0.8314 0.0881 0.0616
0.0041 39.0 6513 0.3445 0.8325 0.2625 1.2894 0.8325 0.8330 0.0880 0.0619
0.0041 40.0 6680 0.3462 0.8317 0.2614 1.2840 0.8317 0.8323 0.0844 0.0599
0.0041 41.0 6847 0.3437 0.833 0.2602 1.2694 0.833 0.8336 0.0871 0.0598
0.003 42.0 7014 0.3456 0.8347 0.2605 1.2867 0.8347 0.8352 0.0844 0.0615
0.003 43.0 7181 0.3454 0.8347 0.2607 1.2844 0.8347 0.8354 0.0868 0.0598
0.003 44.0 7348 0.3451 0.8337 0.2599 1.2719 0.8337 0.8342 0.0832 0.0595
0.0022 45.0 7515 0.3460 0.8343 0.2607 1.2750 0.8343 0.8346 0.0844 0.0591
0.0022 46.0 7682 0.3461 0.8325 0.2607 1.2774 0.8325 0.8328 0.0861 0.0604
0.0022 47.0 7849 0.3465 0.8335 0.2610 1.2776 0.8335 0.8338 0.0834 0.0600
0.0018 48.0 8016 0.3468 0.8327 0.2609 1.2758 0.8327 0.8331 0.0865 0.0605
0.0018 49.0 8183 0.3466 0.8333 0.2610 1.2724 0.8333 0.8336 0.0858 0.0609
0.0018 50.0 8350 0.3468 0.8335 0.2611 1.2696 0.8335 0.8338 0.0865 0.0606

Framework versions