This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 1.0408
Accuracy: 0.9828
F1 Score: 0.9839
Recall: 0.9828

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: 1e-05
train_batch_size: 64
eval_batch_size: 64
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 150
num_epochs: 100
label_smoothing_factor: 0.15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Score	Recall
2.8990	1.0	19	2.8243	0.3	0.1902	0.2394
2.6775	2.0	38	2.5749	0.3759	0.2813	0.3098
2.3450	3.0	57	2.2350	0.5931	0.5678	0.5485
1.9329	4.0	76	1.8672	0.7828	0.7860	0.7689
1.5705	5.0	95	1.5231	0.8931	0.9009	0.8963
1.3488	6.0	114	1.3058	0.9379	0.9421	0.9399
1.1881	7.0	133	1.2164	0.9379	0.9429	0.9400
1.1692	8.0	152	1.1655	0.9483	0.9521	0.9498
1.1013	9.0	171	1.1383	0.9621	0.9654	0.9620
1.0939	10.0	190	1.1244	0.9586	0.9621	0.9570
1.0743	11.0	209	1.1158	0.9690	0.9716	0.9682
1.0535	12.0	228	1.0974	0.9655	0.9687	0.9670
1.0468	13.0	247	1.0882	0.9655	0.9683	0.9683
1.0113	14.0	266	1.0735	0.9759	0.9778	0.9766
1.0041	15.0	285	1.0831	0.9724	0.9753	0.9717
1.0087	16.0	304	1.0676	0.9724	0.9749	0.9754
1.0055	17.0	323	1.0714	0.9655	0.9689	0.9705
0.9963	18.0	342	1.0680	0.9690	0.9719	0.9730
0.9857	19.0	361	1.0630	0.9759	0.9776	0.9753
0.9784	20.0	380	1.0617	0.9724	0.9748	0.9754
0.9790	21.0	399	1.0653	0.9724	0.9748	0.9754
0.9791	22.0	418	1.0574	0.9724	0.9749	0.9754
0.9900	23.0	437	1.0858	0.9690	0.9720	0.9768
0.9790	24.0	456	1.0520	0.9759	0.9776	0.9753
0.9708	25.0	475	1.0499	0.9724	0.9749	0.9754
0.9708	26.0	494	1.0448	0.9793	0.9809	0.9803
0.9707	27.0	513	1.0459	0.9793	0.9809	0.9803
0.9729	28.0	532	1.0549	0.9690	0.9719	0.9730
0.9845	29.0	551	1.0451	0.9793	0.9809	0.9803
0.9771	30.0	570	1.0502	0.9759	0.9778	0.9766
0.9716	31.0	589	1.0507	0.9759	0.9779	0.9779
0.9669	32.0	608	1.0457	0.9793	0.9809	0.9803
0.9761	33.0	627	1.0483	0.9793	0.9808	0.9791
0.9627	34.0	646	1.0437	0.9724	0.9743	0.9743
0.9738	35.0	665	1.0431	0.9759	0.9779	0.9779
0.9666	36.0	684	1.0413	0.9759	0.9779	0.9779
0.9657	37.0	703	1.0452	0.9724	0.9743	0.9743
0.9649	38.0	722	1.0440	0.9793	0.9809	0.9803
0.9607	39.0	741	1.0389	0.9759	0.9774	0.9780
0.9693	40.0	760	1.0408	0.9828	0.9839	0.9828
0.9692	41.0	779	1.0374	0.9793	0.9809	0.9803
0.9599	42.0	798	1.0390	0.9793	0.9809	0.9803
0.9601	43.0	817	1.0372	0.9793	0.9810	0.9816
0.9642	44.0	836	1.0432	0.9793	0.9809	0.9803
0.9624	45.0	855	1.0412	0.9759	0.9779	0.9779
0.9598	46.0	874	1.0395	0.9793	0.9809	0.9803
0.9567	47.0	893	1.0414	0.9793	0.9809	0.9803
0.9634	48.0	912	1.0416	0.9759	0.9779	0.9779
0.9662	49.0	931	1.0416	0.9759	0.9779	0.9779
0.9652	50.0	950	1.0457	0.9759	0.9779	0.9779
0.9597	51.0	969	1.0418	0.9724	0.9748	0.9754
0.9675	52.0	988	1.0403	0.9724	0.9748	0.9754
0.9615	53.0	1007	1.0389	0.9724	0.9749	0.9754

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 5.0.0
Tokenizers 0.22.2

nqvii/fold_3

fold_3

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

nqvii/
fold_3