vit-base-patch16-224-for-pre_evaluation

This model is a fine-tuned version of google/vit-base-patch16-224 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.6048
• Accuracy: 0.3929

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.5774	0.98	16	1.5109	0.3022
1.4794	1.97	32	1.4942	0.3242
1.4536	2.95	48	1.4943	0.3187
1.421	4.0	65	1.4247	0.3407
1.3882	4.98	81	1.4944	0.3462
1.3579	5.97	97	1.4180	0.3571
1.2838	6.95	113	1.4693	0.3681
1.2695	8.0	130	1.4359	0.3434
1.2016	8.98	146	1.4656	0.3599
1.2087	9.97	162	1.4550	0.3379
1.206	10.95	178	1.5056	0.3516
1.1236	12.0	195	1.5003	0.3434
1.0534	12.98	211	1.5193	0.3269
1.0024	13.97	227	1.4890	0.3681
0.9767	14.95	243	1.5628	0.3434
0.9201	16.0	260	1.6306	0.3516
0.9136	16.98	276	1.5715	0.3626
0.8566	17.97	292	1.5966	0.3654
0.8273	18.95	308	1.6048	0.3929
0.7825	20.0	325	1.6175	0.3846
0.736	20.98	341	1.6526	0.3929
0.7008	21.97	357	1.6563	0.3736
0.6714	22.95	373	1.7319	0.3901
0.7039	24.0	390	1.6866	0.3929
0.628	24.98	406	1.7023	0.3791
0.6182	25.97	422	1.7301	0.3901
0.5957	26.95	438	1.7157	0.3846
0.5973	28.0	455	1.7478	0.3709
0.5655	28.98	471	1.7377	0.3736
0.5631	29.54	480	1.7374	0.3736

Framework versions

• Transformers 4.33.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3