vit-base-patch16-224-mascotas

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: 0.5660
• Accuracy: 1.0

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: 5.5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6192	0.99	52	0.5660	1.0
0.2794	2.0	105	0.1524	1.0
0.2006	2.99	157	0.1143	0.9474
0.2106	4.0	210	0.0506	1.0
0.1077	4.99	262	0.0339	1.0
0.1379	6.0	315	0.1391	0.9474
0.1387	6.99	367	0.0711	0.9474
0.1342	8.0	420	0.0231	1.0
0.0803	8.99	472	0.0475	1.0
0.097	10.0	525	0.0190	1.0
0.0888	10.99	577	0.0220	1.0
0.0668	12.0	630	0.0078	1.0
0.0559	12.99	682	0.0073	1.0
0.0759	14.0	735	0.0055	1.0
0.081	14.99	787	0.0058	1.0
0.0806	16.0	840	0.0102	1.0
0.0568	16.99	892	0.0164	1.0
0.0696	18.0	945	0.0049	1.0
0.0692	18.99	997	0.0040	1.0
0.0929	20.0	1050	0.0030	1.0
0.1169	20.99	1102	0.0183	1.0
0.0385	22.0	1155	0.0576	0.9474
0.0564	22.99	1207	0.0512	0.9474
0.0206	24.0	1260	0.0025	1.0
0.0984	24.99	1312	0.0028	1.0
0.0368	26.0	1365	0.0037	1.0
0.0436	26.99	1417	0.0609	0.9474
0.0896	28.0	1470	0.0154	1.0
0.079	28.99	1522	0.0026	1.0
0.0211	30.0	1575	0.0045	1.0
0.0499	30.99	1627	0.0042	1.0
0.0137	32.0	1680	0.0043	1.0
0.0711	32.99	1732	0.0019	1.0
0.0369	34.0	1785	0.0021	1.0
0.0382	34.99	1837	0.0031	1.0
0.0785	36.0	1890	0.0029	1.0
0.0575	36.99	1942	0.0020	1.0
0.0158	38.0	1995	0.0019	1.0
0.0489	38.99	2047	0.0022	1.0
0.0511	39.62	2080	0.0023	1.0

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu118
• Datasets 2.16.1
• Tokenizers 0.15.0