square_run_square_run_second_vote_full_pic_stratified_vgg

This model is a fine-tuned version of timm/vgg19_bn.tv_in1k on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7955
• F1 Macro: 0.5728
• F1 Micro: 0.6818
• F1 Weighted: 0.6688
• Precision Macro: 0.6169
• Precision Micro: 0.6818
• Precision Weighted: 0.7081
• Recall Macro: 0.5833
• Recall Micro: 0.6818
• Recall Weighted: 0.6818
• Accuracy: 0.6818

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
1.3975	1.0	58	1.3856	0.3841	0.5076	0.4636	0.5158	0.5076	0.5772	0.4203	0.5076	0.5076	0.5076
1.3378	2.0	116	1.1751	0.5342	0.6212	0.6245	0.5794	0.6212	0.6773	0.5385	0.6212	0.6212	0.6212
1.2419	3.0	174	1.9097	0.4125	0.5	0.4994	0.4951	0.5	0.6007	0.4193	0.5	0.5	0.5
0.7888	4.0	232	2.3180	0.4220	0.5227	0.5014	0.4837	0.5227	0.5692	0.4493	0.5227	0.5227	0.5227
0.3674	5.0	290	2.3266	0.4577	0.5606	0.5440	0.4758	0.5606	0.5806	0.4905	0.5606	0.5606	0.5606
1.3619	6.0	348	2.4841	0.4506	0.5682	0.5545	0.6116	0.5682	0.7073	0.4684	0.5682	0.5682	0.5682
0.0152	7.0	406	2.3663	0.4958	0.5985	0.5836	0.5062	0.5985	0.5959	0.5128	0.5985	0.5985	0.5985
0.0473	8.0	464	4.0208	0.5310	0.6364	0.6166	0.5555	0.6364	0.6346	0.5377	0.6364	0.6364	0.6364
0.0005	9.0	522	3.3065	0.5385	0.6515	0.6278	0.6027	0.6515	0.6861	0.5596	0.6515	0.6515	0.6515
0.0292	10.0	580	3.3860	0.5622	0.6515	0.6537	0.5888	0.6515	0.6784	0.5595	0.6515	0.6515	0.6515
0.0001	11.0	638	3.5022	0.5771	0.6439	0.6415	0.6222	0.6439	0.6624	0.5665	0.6439	0.6439	0.6439
0.0596	12.0	696	4.5555	0.5334	0.6364	0.6313	0.5599	0.6364	0.6641	0.5455	0.6364	0.6364	0.6364
0.0	13.0	754	4.1290	0.5483	0.6515	0.6450	0.5742	0.6515	0.6685	0.5552	0.6515	0.6515	0.6515
0.0	14.0	812	3.7890	0.6120	0.6894	0.6807	0.6576	0.6894	0.7102	0.6157	0.6894	0.6894	0.6894
0.0001	15.0	870	3.8202	0.5652	0.6591	0.6607	0.6090	0.6591	0.7080	0.5695	0.6591	0.6591	0.6591
0.0	16.0	928	3.6953	0.6044	0.7121	0.6988	0.6256	0.7121	0.7140	0.6132	0.7121	0.7121	0.7121
0.3283	17.0	986	3.5991	0.6040	0.7121	0.6985	0.6185	0.7121	0.7076	0.6129	0.7121	0.7121	0.7121
0.0007	18.0	1044	4.3257	0.5279	0.6364	0.6243	0.5694	0.6364	0.6661	0.5434	0.6364	0.6364	0.6364
0.0	19.0	1102	4.0006	0.5727	0.6742	0.6683	0.5996	0.6742	0.6944	0.5805	0.6742	0.6742	0.6742
0.0	20.0	1160	3.9127	0.5786	0.6818	0.6753	0.5988	0.6818	0.6934	0.5852	0.6818	0.6818	0.6818
0.0	21.0	1218	3.7926	0.5794	0.6894	0.6781	0.5923	0.6894	0.6866	0.5874	0.6894	0.6894	0.6894
0.0	22.0	1276	3.9598	0.5880	0.6894	0.6831	0.6107	0.6894	0.7028	0.5936	0.6894	0.6894	0.6894
0.0	23.0	1334	3.9418	0.5630	0.6742	0.6597	0.5890	0.6742	0.6798	0.5729	0.6742	0.6742	0.6742
0.0001	24.0	1392	3.9533	0.5632	0.6742	0.6607	0.5802	0.6742	0.6698	0.5692	0.6742	0.6742	0.6742
0.0	25.0	1450	3.8715	0.5434	0.6591	0.6421	0.5605	0.6591	0.6516	0.5533	0.6591	0.6591	0.6591
0.0	26.0	1508	3.9542	0.5531	0.6667	0.6510	0.5757	0.6667	0.6674	0.5641	0.6667	0.6667	0.6667
0.0	27.0	1566	3.6644	0.5692	0.6894	0.6670	0.5782	0.6894	0.6708	0.5875	0.6894	0.6894	0.6894
0.0	28.0	1624	3.9299	0.5723	0.6818	0.6693	0.5927	0.6818	0.6886	0.5859	0.6818	0.6818	0.6818
0.0	29.0	1682	3.7600	0.5820	0.6970	0.6809	0.5965	0.6970	0.6910	0.5952	0.6970	0.6970	0.6970
0.0	30.0	1740	3.8283	0.5638	0.6818	0.6636	0.5796	0.6818	0.6752	0.5798	0.6818	0.6818	0.6818

Framework versions

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