WS800_ViT_42895082

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: 0.0776
• Accuracy: 0.9875

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	5	0.6859	0.925
No log	2.0	10	0.6328	0.975
No log	3.0	15	0.5301	0.975
No log	4.0	20	0.4404	0.9625
No log	5.0	25	0.3480	0.975
No log	6.0	30	0.2758	0.975
No log	7.0	35	0.2179	0.9875
No log	8.0	40	0.1789	0.9875
No log	9.0	45	0.1505	0.9875
0.3788	10.0	50	0.1296	0.9875
0.3788	11.0	55	0.1145	0.9875
0.3788	12.0	60	0.1034	0.9875
0.3788	13.0	65	0.0954	0.9875
0.3788	14.0	70	0.0895	0.9875
0.3788	15.0	75	0.0853	0.9875
0.3788	16.0	80	0.0822	0.9875
0.3788	17.0	85	0.0801	0.9875
0.3788	18.0	90	0.0787	0.9875
0.3788	19.0	95	0.0779	0.9875
0.0547	20.0	100	0.0776	0.9875

Framework versions

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