swin-tiny-patch4-window7-224-finetuned-fit

This model is a fine-tuned version of xlagor/swin-tiny-patch4-window7-224-finetuned-fit on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.0711
• Accuracy: 0.9772

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: 120
• eval_batch_size: 120
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 480
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 16

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3727	0.99	62	0.1103	0.9680
0.3551	1.99	125	0.1018	0.9701
0.3258	3.0	188	0.0995	0.9706
0.3008	4.0	251	0.0939	0.9712
0.2896	4.99	313	0.0872	0.9730
0.2612	5.99	376	0.0829	0.9739
0.2275	7.0	439	0.0815	0.9748
0.2358	8.0	502	0.0839	0.9739
0.2191	8.99	564	0.0778	0.9775
0.2096	9.99	627	0.0759	0.9769
0.2063	11.0	690	0.0749	0.9778
0.1916	12.0	753	0.0735	0.9775
0.2002	12.99	815	0.0732	0.9781
0.1905	13.99	878	0.0713	0.9784
0.1835	15.0	941	0.0707	0.9784
0.1949	15.81	992	0.0711	0.9772

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3