vit-base-patch16-224-13_model

This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.5185
• Accuracy: 0.835

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: 0.0002
• 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: 9

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.7535	0.9787	23	1.3773	0.545
0.9606	2.0	47	1.1264	0.625
0.5199	2.9787	70	0.7703	0.705
0.3037	4.0	94	0.6922	0.745
0.1607	4.9787	117	0.5718	0.81
0.148	6.0	141	0.5436	0.82
0.1238	6.9787	164	0.5454	0.805
0.0889	8.0	188	0.5023	0.84
0.0745	8.8085	207	0.5185	0.835

Framework versions

• Transformers 4.40.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1