results

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: 1.4656
• Accuracy: 0.8125
• F1: 0.8141

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	17	0.8876	0.5865	0.5688
No log	2.0	34	0.8620	0.6090	0.6067
No log	3.0	51	0.7611	0.6842	0.6783
No log	4.0	68	0.6987	0.6842	0.6741
No log	5.0	85	0.6540	0.6917	0.6872
No log	6.0	102	0.7933	0.6767	0.6407
No log	7.0	119	0.4766	0.8195	0.8152
No log	8.0	136	0.4624	0.8271	0.8231
No log	9.0	153	0.4528	0.8271	0.8277
No log	10.0	170	0.4641	0.8120	0.8087
No log	11.0	187	0.6063	0.7368	0.7231
No log	12.0	204	0.4783	0.7594	0.7596
No log	13.0	221	0.4987	0.7970	0.7990
No log	14.0	238	0.6023	0.7669	0.7603
No log	15.0	255	0.4588	0.8271	0.8254
No log	16.0	272	0.4362	0.8120	0.8130
No log	17.0	289	0.5342	0.8271	0.8280
No log	18.0	306	0.5012	0.8120	0.8124
No log	19.0	323	0.4891	0.8496	0.8498
No log	20.0	340	0.8525	0.7744	0.7714
No log	21.0	357	0.5291	0.8195	0.8209
No log	22.0	374	0.5355	0.8271	0.8264
No log	23.0	391	0.6323	0.8045	0.8041
No log	24.0	408	0.6973	0.8346	0.8334
No log	25.0	425	0.6705	0.8571	0.8569
No log	26.0	442	0.6056	0.8571	0.8572
No log	27.0	459	0.7864	0.8421	0.8421
No log	28.0	476	0.7067	0.8346	0.8351
No log	29.0	493	0.6695	0.8571	0.8567
0.3504	30.0	510	0.6680	0.8647	0.8646

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0