square_run_second_vote_full_pic_stratified

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

• Loss: 0.8527
• F1 Macro: 0.6046
• F1 Micro: 0.7121
• F1 Weighted: 0.6979
• Precision Macro: 0.6148
• Precision Micro: 0.7121
• Precision Weighted: 0.7042
• Recall Macro: 0.6157
• Recall Micro: 0.7121
• Recall Weighted: 0.7121
• Accuracy: 0.7121

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Macro	F1 Micro	F1 Weighted	Precision Macro	Precision Micro	Precision Weighted	Recall Macro	Recall Micro	Recall Weighted	Accuracy
1.8233	1.0	58	1.8311	0.1942	0.2727	0.2274	0.1804	0.2727	0.2139	0.2367	0.2727	0.2727	0.2727
1.8698	2.0	116	1.9889	0.0840	0.1742	0.0989	0.0796	0.1742	0.1011	0.1719	0.1742	0.1742	0.1742
1.9152	3.0	174	1.7868	0.1549	0.2727	0.1775	0.3078	0.2727	0.3022	0.2068	0.2727	0.2727	0.2727
1.6147	4.0	232	1.9341	0.2055	0.2955	0.2523	0.3212	0.2955	0.3791	0.2645	0.2955	0.2955	0.2955
1.6414	5.0	290	1.4736	0.3384	0.4470	0.3905	0.3487	0.4470	0.3844	0.3667	0.4470	0.4470	0.4470
1.1299	6.0	348	1.2117	0.4394	0.5227	0.5100	0.4779	0.5227	0.5389	0.4409	0.5227	0.5227	0.5227
1.493	7.0	406	1.3023	0.4501	0.5227	0.5130	0.4975	0.5227	0.5811	0.4699	0.5227	0.5227	0.5227
1.191	8.0	464	1.1375	0.5136	0.6136	0.5937	0.5140	0.6136	0.5852	0.5227	0.6136	0.6136	0.6136
1.6657	9.0	522	0.9972	0.5421	0.6439	0.6344	0.5400	0.6439	0.6329	0.5526	0.6439	0.6439	0.6439
0.6272	10.0	580	1.1733	0.4743	0.5833	0.5586	0.4990	0.5833	0.5887	0.4983	0.5833	0.5833	0.5833
0.3887	11.0	638	1.2098	0.4849	0.5833	0.5713	0.4956	0.5833	0.5931	0.5094	0.5833	0.5833	0.5833
0.5232	12.0	696	1.1906	0.5205	0.6212	0.6061	0.5470	0.6212	0.6370	0.5378	0.6212	0.6212	0.6212
1.1531	13.0	754	1.1958	0.5960	0.6439	0.6364	0.6959	0.6439	0.6619	0.5765	0.6439	0.6439	0.6439
0.4566	14.0	812	1.2707	0.5381	0.6061	0.5919	0.5768	0.6061	0.6055	0.5356	0.6061	0.6061	0.6061
0.6865	15.0	870	1.3936	0.5478	0.6364	0.6304	0.5668	0.6364	0.6634	0.5658	0.6364	0.6364	0.6364
0.4421	16.0	928	1.3080	0.5732	0.6591	0.6616	0.5855	0.6591	0.6781	0.5718	0.6591	0.6591	0.6591
0.3093	17.0	986	1.7662	0.4753	0.5682	0.5524	0.5212	0.5682	0.6010	0.4840	0.5682	0.5682	0.5682
0.134	18.0	1044	1.4813	0.5427	0.6364	0.6250	0.5700	0.6364	0.6476	0.5510	0.6364	0.6364	0.6364
0.0406	19.0	1102	1.5180	0.5583	0.6515	0.6354	0.5572	0.6515	0.6338	0.5736	0.6515	0.6515	0.6515
0.0198	20.0	1160	1.8253	0.5403	0.6212	0.6226	0.5579	0.6212	0.6497	0.5487	0.6212	0.6212	0.6212
0.0054	21.0	1218	1.6285	0.5574	0.6515	0.6478	0.5722	0.6515	0.6595	0.5596	0.6515	0.6515	0.6515
0.2816	22.0	1276	1.7743	0.5174	0.6212	0.6028	0.5400	0.6212	0.6203	0.5257	0.6212	0.6212	0.6212
0.0076	23.0	1334	1.7558	0.5353	0.6212	0.6174	0.5506	0.6212	0.6324	0.5389	0.6212	0.6212	0.6212
0.0046	24.0	1392	1.7770	0.5518	0.6364	0.6346	0.5720	0.6364	0.6581	0.5580	0.6364	0.6364	0.6364
0.0018	25.0	1450	1.5917	0.5926	0.6818	0.6812	0.6030	0.6818	0.6864	0.5869	0.6818	0.6818	0.6818
0.0041	26.0	1508	1.7247	0.5866	0.6667	0.6725	0.6022	0.6667	0.6874	0.5788	0.6667	0.6667	0.6667
0.0013	27.0	1566	1.6674	0.5977	0.6894	0.6852	0.6010	0.6894	0.6861	0.5997	0.6894	0.6894	0.6894
0.0022	28.0	1624	1.7056	0.5938	0.6818	0.6794	0.5964	0.6818	0.6828	0.5975	0.6818	0.6818	0.6818
0.001	29.0	1682	1.6834	0.5978	0.6894	0.6849	0.5975	0.6894	0.6828	0.6006	0.6894	0.6894	0.6894
0.0006	30.0	1740	1.6730	0.6075	0.6970	0.6932	0.6149	0.6970	0.6968	0.6081	0.6970	0.6970	0.6970

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu124
• Datasets 3.3.1
• Tokenizers 0.21.0