vit-tiny-patch16-224

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.6141
• F1 Macro: 0.4385
• F1 Micro: 0.5303
• F1 Weighted: 0.4856
• Precision Macro: 0.5225
• Precision Micro: 0.5303
• Precision Weighted: 0.5788
• Recall Macro: 0.4858
• Recall Micro: 0.5303
• Recall Weighted: 0.5303
• Accuracy: 0.5303

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
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH 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: 25

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
1.9719	1.0	29	1.9209	0.0949	0.2197	0.1155	0.0891	0.2197	0.1051	0.1722	0.2197	0.2197	0.2197
1.8717	2.0	58	2.0378	0.0953	0.1970	0.1069	0.1996	0.1970	0.2660	0.1794	0.1970	0.1970	0.1970
1.9326	3.0	87	1.7680	0.2290	0.3939	0.2939	0.2151	0.3939	0.2682	0.3004	0.3939	0.3939	0.3939
1.2873	4.0	116	1.5892	0.3502	0.4470	0.4082	0.4831	0.4470	0.5140	0.3646	0.4470	0.4470	0.4470
1.3997	5.0	145	1.4773	0.3481	0.5	0.4245	0.3463	0.5	0.4119	0.4052	0.5	0.5	0.5
1.7041	6.0	174	1.4406	0.4266	0.5379	0.5005	0.5011	0.5379	0.5628	0.4529	0.5379	0.5379	0.5379
1.1863	7.0	203	1.3680	0.4759	0.5682	0.5400	0.5559	0.5682	0.6032	0.4831	0.5682	0.5682	0.5682
0.9817	8.0	232	1.3515	0.4399	0.5227	0.4969	0.4445	0.5227	0.5088	0.4722	0.5227	0.5227	0.5227
0.617	9.0	261	1.3867	0.4895	0.5909	0.5555	0.5136	0.5909	0.5776	0.5183	0.5909	0.5909	0.5909
1.0365	10.0	290	1.4607	0.4313	0.5379	0.4961	0.4371	0.5379	0.4997	0.4674	0.5379	0.5379	0.5379
0.6815	11.0	319	1.3133	0.4962	0.5909	0.5664	0.5087	0.5909	0.5742	0.5133	0.5909	0.5909	0.5909
0.4153	12.0	348	1.3528	0.5082	0.5909	0.5735	0.5185	0.5909	0.5820	0.5202	0.5909	0.5909	0.5909
0.3396	13.0	377	1.3856	0.5372	0.5909	0.5830	0.5623	0.5909	0.6018	0.5387	0.5909	0.5909	0.5909
0.5415	14.0	406	1.4252	0.5132	0.5909	0.5795	0.5223	0.5909	0.5893	0.5255	0.5909	0.5909	0.5909
0.4421	15.0	435	1.4081	0.5574	0.6136	0.6086	0.5753	0.6136	0.6149	0.5532	0.6136	0.6136	0.6136
0.2893	16.0	464	1.5285	0.5127	0.5985	0.5833	0.5059	0.5985	0.5752	0.5253	0.5985	0.5985	0.5985
0.2403	17.0	493	1.4820	0.5395	0.6288	0.6065	0.5808	0.6288	0.6380	0.5460	0.6288	0.6288	0.6288
0.1087	18.0	522	1.3999	0.5320	0.6061	0.6009	0.5612	0.6061	0.6211	0.5261	0.6061	0.6061	0.6061
0.2619	19.0	551	1.4408	0.5618	0.6136	0.6037	0.6154	0.6136	0.6225	0.5501	0.6136	0.6136	0.6136
0.1154	20.0	580	1.4516	0.5402	0.6288	0.6090	0.5538	0.6288	0.6145	0.5492	0.6288	0.6288	0.6288
0.1367	21.0	609	1.5306	0.5254	0.6136	0.5942	0.5321	0.6136	0.5923	0.5340	0.6136	0.6136	0.6136
0.0839	22.0	638	1.6397	0.5154	0.5833	0.5756	0.5274	0.5833	0.5895	0.5252	0.5833	0.5833	0.5833
0.1818	23.0	667	1.6416	0.5656	0.6515	0.6359	0.5848	0.6515	0.6456	0.5696	0.6515	0.6515	0.6515
0.0781	24.0	696	1.6026	0.5393	0.6212	0.6079	0.5524	0.6212	0.6118	0.5412	0.6212	0.6212	0.6212
0.0792	25.0	725	1.5997	0.5494	0.6288	0.6180	0.5716	0.6288	0.6297	0.5480	0.6288	0.6288	0.6288

Framework versions

• Transformers 4.48.2
• Pytorch 2.6.0+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0